(navigation image)
Home American Libraries | Canadian Libraries | Universal Library | Community Texts | Project Gutenberg | Biodiversity Heritage Library | Children's Library | Additional Collections
Search: Advanced Search
Anonymous User (login or join us)
See other formats

Full text of "An introduction to natural philosophy designed as a text-book for the use of students in college"


















Entered according to Act of Congress, in the year 1844, by 

In the Clerk s Office of the District Court of Connecticut. 

Entered according to Act of Congress, in the year 1860, by 


In the Clerk s Office of the District Court of the District of Connecticut. 

Entered according to Act of Congress, in the year 1870, by 


In the Office of the Librarian of Congress, at Washington. 




Electrotyped by SMITH & MCDOUGAL. 82 Beekman Street. 


THE object kept in view in the present revision is the same 
as heretofore to prepare a book suitable for use in the 
College recitation- room. The work does not aim to be a hand 
book of Physics, giving information on all points relating to the 
subjects treated of, but merely a book of first principles, accom 
panied by a sufficient number of illustrative statements and dia 
grams to render those principles clear, and to impress them on 
the memory. 

The more prominent departures from the former revision are 
the following: 

1. Most of the second part of Mechanics is omitted; and what 
is retained is introduced in appropriate connections in Part I. 
The second part was originally intended as a sort of substitute 
for a course of experimental lectures. But colleges are now so 
generally supplied with apparatus for illustration, that it seems 
unnecessary to encumber the volume with information which can 
be presented so much more satisfactorily by the lecturer. 

2. Instead of the brief Part entitled "Electro-magnetism," 
first presented in the work when the former revision was made, 
the subject of " Dynamical Electricity" is now discussed as fully 
as the other branches. 

3. The subject of "Heat" for the first time forms one Part of 
the work, although the course of instruction in many colleges 
still retains it in the chemical department. 

4. Some additions are made to the applications of the differ 
ential and integral calculus, and all the discussions of this char 
acter are brought together in an " Appendix " at the close of the 

5. More than three-fourths of the engravings are new, most 
of which were drawn expressly for this revision. 



Besides these more apparent alterations, it should be addt-d 
that a large part of the whole book has been carefully rewritten, 
and additions and improvements made in almost every page of it. 

The author of the revision wishes to express his indebtedness 
to Professor JOSEPH EICKLIK, of the University of Missouri, who 
has furnished much valuable material for the Part on " Mechan 
ics," and has critically examined the mathematical portions of the 
book, and rendered essential aid in correcting and improving 
them. The Part on " Dynamical Electricity " is almost entirely 
the work of Professor CHAKLES H. SMITH, of Cincinnati, Ohio, 
formerly Tutor of Natural Philosophy in Yale College. Much 
credit is due to him for presenting the principles of that exten 
sive and complex branch of physics in a clear and systematic 
form. Dr. B. JOY JEFFRIES, of Boston, Lecturer on Optical Phe 
nomena and the Eye in Harvard University, has kindly assisted 
in preparing the description of the eye and its adjustments, and 
has allowed his own original drawings to be copied in the engrav 
ings relating to this subject. 


AMHERST COLLEGE, September, 1870. 




Classification of the Physical Sciences. Definitions relating to matter. 
Properties of matter. Branches of Natural Philosophy 1 3 



Motions classified. Uniform motion. Momentum. Forces classified. 
Laws of motion. Gravity. Its laws. Questions 4 12 


Uniform and variable motion represented geometrically. Laws of the 
fall of a body. Motion of a body projected up or down. Formulae for 
the fall of bodies. Space and time represented by co-ordinates. At- 
wood s machine. Living force. Questions 12 23 


Two or more forces acting on a body. Parallelogram of forces. Triangle 
of forces. Polygon of "forces. Curvilinear motion. Calculation of the 
resultant. Examples. Resolution of motion. Resultant found by 
rectangular axes. Analytical expression for the resultant Principle 
of moments. Parallel forces. Parallelopiped of forces. Rectangular 
axes. Equilibrium of forces. Couples. Forces resisted by a smooth 
plane 23 12 


Centre of gravity defined. Centre of two equal bodies Of two unequal 
bodies. Equal moments. Three or more bodies. Triangle. Poly 
gon. Perimeters. Pyramid. Examples Centre of gravity referred 
to a point or plane. Trapezoid. Centrobaric mensuration. Stable, 
unstable, and neutral equilibrium. Motion of the centre of gravity of 
a system. When one body moves, and when more than one move. 
Examples 42 58 


Collision of inelastic bodies. Formulae. Questions. Collision of elastic 
bodies, equal, unequal. Series of bodies, equal, decreasing 1 , increas 
ing. Living force lost in the collision of inelastic forces. Preserved in 
the collision of elastic bodies. Impact on a plane 58 65 


Machines classified. Three orders of straight lever. Equal moments. 
Compound lever. Balance. Steelyard. Platform scales. Wheel and 
axle. The same compounded. Connection by teeth and by bands. 
Pulley. Fixed, movable, compound. Rope machine. Branching 
rope. Funicular polygon. Inclined plane. Relation of power, weight, 
and pressure. Body between two planes. Equilibrium of bodies on 
two planes. Screw. Combined with the lever. Endless screw Right 
and left hand screw. Wedge. Knee-joint. Principle of virtual veloci 
ties. Friction in machinery. Laws of sliding friction. Friction of 
3. Rolling friction. Friction wheels 65 95 



The fraction of gravity on an inclined plane. Formulas for descent. 
Descent on the chords of a circle. Series of planes. Descent, on a 
curve. The pendulum. Calculation of its length. Point of suspen 
sion and centre of oscillation interchangeable. The cycloid. Its prop 
erties. Descent on a cycloid. The involute of a semi-cycloid. The 
cycloidal pendulum. Results applied to common pendulum. Compen 
sation pendulum 95 108 


Formulge for the path of a projectile. Different angles of elevation for 
the same range. The greatest height of a projectile Formula for a 
horizontal plane. Equation of the path of a projectile. Range on an 
oblique plane. Questions. Central forces described Expressions for 
centrifugal force in circular motion. Two bodies revolving about their 
centre of gravity. Centrifugal force on the earth. Examples. Com 
position of two rotary motions The gyroscope 109 120 


Longitudinal strength. Lateral strength, both ends supported. Stress 
from weight of the beam. Weight at the centre of the beam. 
Weight at any other point. Form for equal strength. Lateral 
strength, one end supported. Prismatic beams breaking by their own 
weight. Structures weaker as they are larger. Solid and hollow cyl 
inders. Miscellaneous problems in Mechanics 120 132 



Liquids distinguished from solids and gases. Transmitted pressure. 
Hydraulic press. Equilibrium of a liquid. Curvature of surface. 
Spirit level Pressure as depth. Amount of pressure. Artesian 
wells Centre of pressure Loss of weight in water. Equilibrium 
of floating bodies Specific gravity Hydrometer. Magnitude found 
by specific gravity. Cohesion and adhesion. Capillary action. 
Liquids raised and depressed by the side of a solid. Capillary tubes 
and plates. Effect of capillarity on floating bodies 133 152 


Depth and velocity of discharge. Descent of surface Orifices in differ 
ent situations. Vena contracta, Friction in pipes. Jets. Rivers. 
Lifting pump. Chain pump. Hydraulic ram. Water- wheels. Tur 
bine. Barker s mill. Resistance of a liquid. Waves of oscillation. 
Molecular movements. Sea waves. Waves of translation 153 166 



Nature of gases. Mariotte s law. The air-pump. Rate of exhaustion. 
Air condenser. Torricelli s experiment. Atmospheric pressure. Ba- 
rometer.-s-Pressure at different latitudes. Diurnal variation. Effect 
of weather on pressure. Heights measured by the barometer. Gauges 
of the air-pump ; 167 176 


Bellows. Siphon, Suction pump. Forcing-pump. Fire-engine 
Hero s fountain. Manometer. Apparatus for preserving a water- 
level . 177182 



Quantity of the atmosphere. Its height. Change of density with 
height. Trade winds. Return currents. Wind of higher latitudes. 
Land and sea breezes. Current in a medium. Effect on a surface. 
Vortices 182188 



Vibrations the cause of sound. Sonorous bodies. Air the common me 
dium. Velocity in air. Diffusion of sound. Nature of the waves. 
Gases, liquids, and solids as media of sound. Mixed media 189 196 


Laws of reflection of sound. Echoes. Concentrated echoes. Resonance 
of rooms. Halls for public speaking. Refraction of sound. Inflection 
of sound 197202 


Vibrations in musical sounds. Pitch. The monochord. Formula for 
time of a vibration The number of vibrations in a given time. Vi 
brations of a column of air. Modes of vibration in different pipes. 
Vibration in parts. Modes of exciting the vibrations. Rods and 
laminae. Chladni s plates. Bells. The voice. The organ of hear 
ing 202214 


Numerical relations of musical sounds. Repetition of the scale. Modes 
of naming the notes. Diatoni: Jind chromatic scales. Chords and dis 
cords. Temperament. Harmonics Overtones. Effect on quality of 
tone . Communication of vibrations . Crispations . Interference . 
Number and length of waves for each note. Vibrations visibly pro 
jected 214-223 



Natural and artificial magnet. Attraction of iron. Polarity. Mutual 
action of magnets. Magnetic induction. Reflex influence. Double 
induction. Coercive force. Magnetism not transferred. Law of 
force and distance. Positions of a needle near a magnet. Magnetic 
curves 224231 


Declination of the needle. Isogonic curves Variations, secular, annual, 
and diurnal. Dip of the needle. Isoclinic curves. Magnetic inten 
sity. Isodynamic curves. Magnetic observatories. Aurora Borealis. 
Source of the earth s magnetism. Formation of permanent magnets. 
The declination compass. The mariner s compass. The needle ren 
dered astatic. Theory of magnetism 231 241 


Definitions. Electroscopes. Two electrical states . Theories.-*-Mutnal 
action. Conduction. Modes of insulating Sphere of communication 
and of influence 242247 



Plate machine. Cylinder machine. Hydro-electric machine. Phenom 
ena of the machine. Torsion balance. Law of force. Charge on the 
surface. How distributed. Held by the air 247253 


Elementary experiment on induction. Reaction. Effect of dividing the 
conductor. Of lengthening it. Disguised electricity. Series of con 
ductors. Why an unelectrified body is always attracted. The Franklin 
plate. The Leyden jar. Its theory. Spontaneous discharge. Series 
of jars. Dividing a charge. Use of coatings. The free part of the 
charge explained. Vibrations and revolutions. Residuary discharge. 
Battery. Discharging electrometers. Why a point held toward a 
charged body discharges it. Induction applied to the explanation of 
electroscopes, condenser, and electrophorus. Induction machine.. 253 266 


Effects of discharges. Luminous effects. Colors. Luminous figures. 
Mechanical, chemical, and physiological effects. Velocity of Elec 
tricity 266270 


Electricity in the air. Thunder storms. Lightning is a discharge of 
electricity. Rods. How they protect. Protection of the person. 
How lightning causes damage 271 275 


Electricity by chemical action. An element. A battery. Comparison 
of statical and dynamical electricity. Quantity batteries, intensity bat 
teries 2? 6280 


Helices. The solenoid. Ampere s theory Mutual action of currents. 
Relations of currents and magnets. The galvanometer. The earth s 
polarity. Thermo-electricity. Magnetic induction by currents. The 
U-magnet. Its power to sustain a weight 280 288 


Currents inducing currents. Characteristics of induced currents. Both 
currents in one wire. Names of these circuits and currents. Coils, 
primary and secondary. Ruhmkorff s coil. One coil moved into and 
out of another. Magneto-electricity. ^Explained on Ampere s theory. 
Clarke s magneto-electric machine. Its operation 289 298 


Practical applications. Electrolysis . Electro-plating. Electrotyping. 
Electric light and heat. Mechanical movements. Electro-magnetic 
machine. Electro-magnetic telegraph. Its parts. Its operation. Re 
peaters. Ocean cable. Fire-alarm. Chronograph 298 309 



Nature of h6at, Expansion by heat, contraction by its loss. Thermom 
eter. Different scales. Pyrometer. Coefficient of expansion. Ther 
mal force very great. Case of contraction by heat 310 315 



Heat communicated in several ways. Radiation. Heat tends to an equi 
librium. Reflection. Concentration by reflection. Absorption. Con 
duction. Effect of molecular arrangement in solids. Convection. 
Diathermancy 315820 


Specific heat. Method of finding it. Change of condition. Latent 
heat. Boiling under pressure. Freezing by melting. Spheroidal con 
dition 321 325 


Force of steam. Change of tension with temperature. Steam-engines. 
The engine of Watt. Single and double acting. Low-pressure en 
gine. Steam-valves. High-pressure engine. Applications of steam- 
power. Mechanical equivalent of heat 325 381 


How the air is warmed. Limit of perpetual frost. Isothermal lines. 
Moisture of the air. Dew point. Measure of vapor. Hygrometers. 
Forms of precipitation of vapor. Dew, frost. Fog, cloud. Classifica 
tion of clouds. Rain, mist, hail, sleet, snow. Theories of precipita 
tion. Cyclones. Draft of flues. Ventilation of rooms, of mines. 
Sources of heat 331 341 



Light moves in straight lines. Its velocity. Intensity at different dis 
tances. Loss by absorption. Photometers. Shadows 342 346 


Reflection. Its law. Inclination of rays not changed by a plane mir 
ror. Rays converged by a concave, diverged by a convex mirror. 
Conjugate foci. Images by a plane mirror. Object and image symmet 
rical. Space on the mirror occupied by the image. Displacement by 
two reflections. Multiplied images by two mirrors, parallel, inclined. 
Images by a concave mirror, by a convex mirror. Caustics by reflec 
tion. Spherical aberration of mirrors 346 3GO 


Refraction. Law as to density. Law as to inclination. Limit of emer 
gence from a denser medium. Transmission through plane surfaces, 
parallel, inclined. Multiplying glass. Refractive power found by a 
prism. Light through one surface, plane, convex, concave. Lenses. 
Effect of the convex lens, of the concave lens. Optic centre. Conju 
gate foci. Images by the convex lens, by the concave lens. Caustics 
by refraction. Spherical aberration of lenses. Remedy. Atmospheric 
refraction. Mirage 300 374 


The prismatic spectrum. Colors recombined. Complementary colors. 
Fraunhofer lines. Lines of terrestrial substances burning. Theory of 
lines in the solar spectrum. Dispersion of light. Chromatic aberra 
tion of lenses. 1 Achromatism 375 381 


The rainbow. Experiment with a sphere of water. Course of rays in 
the primary bow. In the secondary bow. Axis of the bows. Their 
circular form. Colors of the two bows in contrary order. The tertiary 
bow. The common halo. Caused by crystals of ice. Its frequency. 
The mock-sun 381-388 



Natural colors of bodies. Inflection of light. Breadth of fringe varies 
with the color. Why not always seen on the edges of bodies. Color 
by striation. By thin laminae. Ratio of thicknesses for the successive 
rings. Mode of finding the thickness. Newton s rings by a mono 
chromatic lamp 388393 


Double refraction. Iceland spar. Ordinary and extraordinary ray. 
Optical relations of the axis. Crystals of positive and of negative 
axis. Polarization of light. By reflection. Polarizing and analyzing 
plates. By bundle of plates. By absorption. By double refraction. 
Every polarizer an analyzer. Color by polarized light 393 399 


The wave theory. Its postulates. Reflection according to it. Refrac 
tion according to each theory. Interference. By thin plates. Bj two 
mirrors. By inflection. Length of waves and number per second for 
each color. Moc^e of vibration in polarized light. Application in the 
several modes of polarizing 399 407 


Image by light through an aperture. Effect of a convex lens at the aper 
ture. The eye. Parts of the interior. Vision. Adaptations. Ac 
commodation. How caused. Long-sightedness. Short-sightedness. 
Cause of each. Why an object is seen erect and single. ndirect 
vision. The blind point, Continuance of impressions. Accidental 
colors. Estimate of distance and size by the eye. Binocular vision. 
The stereoscope 407 415 


The camera lucida. The microscope. The single microscope. Limit of 
its power. The compound microscope. Its magnifying power. Im 
provements in its construction. Microscopes for projecting images. 
The magic lantern. The solar microscope. The astronomical tele 
scope. Its powers. Mode of mounting. The terrestrial telescope. 
Galileo s telescope. The Gregorian telescope. The Herschelian tele 
scope 416434 



Differential equations. Fall through small distances near the earth. 
Through great distances. Method of finding velocity and time. Fall 
within the earth. Velocity and time found 425428 


Principle of moments. Formula prepared. Applications of formulae to 
various cases 428 432 


Moment of inertia for any axis. Examples 432 434 


General formulae. Examples 434436 


The primary bow. The secondary bow. The halo 436437 



Art. 1. Classification of Physical Sciences. The ma 
terial world consists of two parts the organized, including the 
animal and vegetable kingdoms; and the unorganized, which 
comprehends the remainder. Organized matter is treated of in 
Physiology, and in those branches of science usually called Natural 
History. Unorganized matter forms the subject of Natural Phi 
losophy and Chemistry. Chemistry considers the internal consti 
tution of bodies, and the relations of their smallest parts to each 
other. Natural Philosophy deals principally with the external 
relations of bodies and their action upon one another. If, how 
ever, the bodies are so large as to constitute ivorlds, of which the 
earth itself is one, this science takes the name of Astronomy. 

The word Physics is much used to include both Natural 
Philosophy and Chemistry; but sometimes it is applied to the 
branches of Natural Philosophy, except Mechanics. According to 
this use of the word, Natural Philosophy is divided into two gen 
eral subjects, Mechanics and Physics. 

2. Definitions relating to Matter. 

A body is a separate portion of matter, whether large or 

An atom is a portion of matter so small as to be indivisible. 

A particle denotes the smallest portion which can result 
from division by mechanical means, and consists of many atoms 
united together. 

The word molecule signifies a very small portion of matter, 
either atom or particle. 

Mass is the quantity of matter in a body, and is usually 
measured by its weight. 

Volume signifies the space occupied by a body. 

Density expresses the relative mass contained within a given 


volume. ,Tlm,:if toiie body has twice as great a mass within a ccr- 
taiVvoltfrrie aVaA6the t r has, it is said to have twice the density. 
; ; ^ Po^es .a^the o^nule portions of space within the volume of 
: a*bedy^ which l are not filled by the material of that body. All 
matter is porous, some kinds in a greater and some in a less degree. 
Force is the name of any cause, whatever it may be, which 
gives motion to matter, or which changes its motion. 

3. Properties of Matter. 

(1.) Extension. Every portion of matter, however small, has 
length, breadth, and thickness, and thus occupies space. This 
is its extension. 

(2.) Impenetrability. While matter occupies space, it. ex 
cludes all other matter from it, so that no two atoms can be in 
exactly the same place at the same time. This property is called 

The two foregoing are often called essential properties, because 
we cannot conceive matter to exist without them. 

(3.) Divisibility. Matter is divisible beyond any known limits. 
After being divided, as far as possible, into particles by mechani 
cal methods, it may be still further reduced by chemical action to 
atoms, which are too small to be in any way recognized by the 

(4.) Compressibility. Since pores exist in all matter, it may 
be compressed into a smaller volume. Hence all matter is com 
pressible, though in very different degrees. 

(5.) Elasticity. After a body has suffered compression, it 
shows, in some degree at least, a tendency to restore itself to its 
former volume. This property is called elasticity. A body is said 
to be perfectly elastic when the force by which it recovers its size 
is equal to that by which it was before compressed. The word 
elasticity is used generally in a wider sense than is given in the 
above definition, namely, the tendency which a body has to recover 
its original form, whatever change of form it may have previously 
received. Thus, if a body is stretched, bent, twisted, or distorted 
in any other way, it is called elastic, if it tends to resume its form 
as soon as the force which altered it has ceased. Torsion is the 
name of the elastic force which tends to untwist a thread or wire 
when it has been twisted. 

(6.) Attraction. This is the general name used to express the 
universal tendency of one portion of matter towards another. It 
receives different names, according to the circumstances in which 
it acts. The attraction which binds together atoms of different 
kinds, so as to form a new substance, is called affinity, and is dis 
cussed in Chemistry; that which unites particles, whether simple 


or compound, so as to form a body, is called cohesion; the cling 
ing of two kinds of matter to each other, without forming a new 
substance, is called adhesion; and the tendency manifested by 
masses of matter toward each other, when at sensible distances, is 
called gravity. 

(7.) Inertia. This is also a universal property of matter, and 
signifies its tendency to continue in its present condition as to 
motion or rest. If at rest, it cannot move itself; if in motion, it 
cannot stop itself or change its motion, either in respect to direc 
tion or velocity. 

4. Branches of Natural Philosophy. Natural Philosophy 
is generally divided into Mechanics, Hydrostatics, Pneumatics, 
Sound, Magnetism, Electricity, Heat, and Light. 

Mechanics treats of the motion and equilibrium of bodies, 
caused by the application of force. Since there are three condi 
tions of matter, solid, liquid, and gaseous, it is convenient to 
divide the general subject of Mechanics into three branches. 

1st. The mechanics of solids, also called Mechanics. 

2d. The mechanics of liquids, called Hydrostatics. 

3d. The mechanics of gases, called Pneumatics. 

All the other branches of Natural Philosophy (often called 
Physics) treat of various phenomena caused by minute vibrations 
in the particles of matter. These vibrations are excited in differ 
ent ways, and when transmitted to us, affect one or more of our 
senses. Thus, sound consists of such vibrations as affect the sense 
of hearing ; and light is another mode of vibration, that affects 
only the sense of vision. 

It was formerly customary to regard magnetism, electricity, 
heat, and light, as so many kinds of imponderable matter, that is, 
matter having no sensible weight, and thus distinguished from 
solids, liquids, and gases, which are the different forms of ponder 
able matter. But it is now known that when forces are applied to 
matter, they not only produce the visible forms of motion, but 
may be made to develop either sound, magnetism, electricity, heat, 
or light; and that most of these modes of motion may be trans 
formed into others, and each may be made a measure of the force 
which is employed to produce it. 





5. Classification of Motions. Motion is change of place, 
and is either uniform or variable. In uniform motion equal 
spaces are passed over in equal times, however small the times may 
be. In variable motion the spaces described in equal times are un 
equal. Such motion may be either accelerated or retarded. In 
accelerated motion the spaces described in equal times become con 
tinually greater; in retarded motion they become continually less. 
Motion is said to be uniformly accelerated if the increments of 
space in equal times (however small) are equal ; and uniformly re 
tarded if the decrements are equal. 

Velocity is the space described in the unit of time. In Me 
chanics, one second is much used as the unit of time, and one foot 
as the unit of space ; hence, velocity is the number of feet de 
scribed in one second. 

6. Uniform Motion. When motion is uniform, the number 
of feet described in one second, multiplied by the number of 
seconds, obviously gives the whole space. Let s = space, t = time, 

n o 

and v = velocity; then s = t v } /. = -, and v = j. If this 

space is compared with another, s , described in the time t , with 
the velocity v , then s : s : : tv : t v ; or briefly, in the form of a 

o o 

variation, s oc t v. In like manner t cc -, and v cc -. 

v t 

If two bodies, moving uniformly, describe equal spaces, then 
s = s ; /. t v = t v ; /. t : t ::v :v. That is, in order that two 
bodies may describe equal spaces, their velocities must vary in 
versely as the times during which they move. 

7. Questions on Uniform Motion. 

1. A ball was rolled on the ice with a velocity of 78 feet per 
second, and moved uniformly 21 seconds; what space did it de 
scribe? Ans. 1G38 feet. 


"2. A steamboat moved uniformly across a lake 17 miles wide, 
at the rate of 20 feet per second; what time was occupied in 
crossing? Ans. Ih. 14ra. 4s. 

3. On the supposition that the earth describes an orbit of GOO 
millions of miles in 365 j days, with what velocity does it move per 
second? Ans. 19 miles, nearly. 

4. Three planets describe orbits which are to each other as 15 
19, and 12, in times which are as 7, 3, and 5 ; what are their rela 
tive velocities ? Ans. 225, 665. and 252. 

8. Momentum. The momentum of a body signifies its quan 
tity of motion, and is reckoned according to the mass, or quantity 
of matter, which is moving, and the velocity with which it moves. 
The momentum, therefore, varies as the product of the mass and 
the velocity. 

Let the momentum of a body = m, its mass q, and its vclo- 

., m m T 

city = v\ then m = q v, q = , and v = -. In order to com 
pare the momentum of one body with that of another, let m , q , v f , 
represent the momentum, mass, and velocity, respectively of the 

second body; then m : m : : q v : q v ; or m oc q V, .*. q cc , 


and v cc . 

If the momentum of one body equals that of another, then, 
since m m , q v = q v , :. q . q :: v : v. That is, in order that 
the momenta of two bodies should be equal, their masses must 
vary inversely as their velocities. 

Since there are two elements entering into the momentum of a 
body namely, its mass, usually expressed in pounds, and its velo 
city, expressed in feet per second therefore momentum cannot be 
measured either in pounds or in feet, being in nature unlike either. 
The word foot-pound is employed for the unit of momentum when 
ever the unit of mass is a pound and the unit of velocity is a foot 
per second. 4 

9. Questions on Momentum. 

1. A ship weighing 336,000 Ibs. is dashed against the rocks in 23 J 
a storm, with a velocity of 16 miles per hour; with what momen- 
turn did she strike? Ans. 7,884,800 foot-pounds. r 

2. A ball weighing 1 oz. is fired into a log weighing 53 Ibs., , 
suspended so as to move freely, and imparts a velocity of 2 ft. per 
second. Assuming that the log and ball have a momentum equal V* 
to the previous momentum of the ball alone, required the velocity 

of the ball. Ans. 1,698 ft. per sec. 

8. Suppose a comet, whose velocity is 1,000,000 miles per hour, 


has the same momentum as the earth, whose velocity is 19 miles 
per second ; what is the ratio of their masses ? Ans. 1 : 14.6. 

4. Two railway cars have their quantities of matter as 7 to 3, 
and their momenta as 8 to 5 ; what are their relative velocities ? 

Ans. As 24 to 35, or nearly 5 to 7. 

5. The momentum of a cannon-ball was 434 foot-pounds ; what 
must be the velocity of a half-ounce bullet, in order to have the 
same momentum ? Ans. 13,888 feet. 

10. Classification of Forces. The principal forces in na 
ture are the following : 

1. Attraction in its several forms. Cohesion and chemical af 
finity are the forces which bind together the particles and atoms 
of bodies, and gravity is that which everywhere near the earth 
causes bodies to fall toward it, or to press upon it. 

2. Elasticity. This is a force which, in many kinds and con 
ditions of matter, tends to repel the particles from each other. 

The forces, whether attraction or repulsion, which exist among 
the atoms or molecules of a body, are called molecular forces. 

3. Muscular force. All living beings are endowed with this 
force, by which they put in motion bodies around them, and by 
acting upon other bodies, are enabled also to move themselves 
from place to place. 

4. Matter in motion. If a body which some force has put in 
motion impinges on another body, it imparts motion to it, and is 
therefore itself a force. This is true not only of ordinary visible 
motions, but of those small and often invisible vibrations, which 
manifest themselves as sound, heat, &c. Gravity, or any other 
force, may cause heat, and heat may cause light and electricity. 
Thus, any form of motion is a force, and it can be employed to 
produce other forms. 

11. Impulsive and Continued Forces and their Ef 
fects. An impulsive force is one which has no sensible continu 
ance, as the blow of a hammer. A continued force is one which 
acts during a perceptible length of time. Continued forces are 
subdivided into constant and variable. A constant force has the 
same intensity during the whole time of its action ; a variable 
force is one whose intensity changes. 

Keeping in mind the property of inertia, we associate different 
kinds of motion with the forces which produce them, as follows : 

1. An impulsive force causes uniform motion. 

2. A continued force, accelerated motion. 

3. A constant force, uniformly accelerated motion. 

4. A variable force, unequally accelerated motion. 

If the force is applied in a direction opposite to that in which 


the body has a previous uniform motion, the conn3ction is the 
following : 

5. An impulsive force causes uniform motion, or rest. 

6. A continued force, retarded motion. 

7. A constant force, uniformly retarded motion. 

8. A variable force, unequally retarded motion. 

In cases 1 and 5, it is obvious that, the impulse being given, 
the body is left to itself, and cannot change the state of motion or 
rest impressed on it 

In 2, 3, and 4, it must be considered that the force at each in 
stant adds a new increment to the uniform motion which the body 
would have had if the force had ceased; and if the force is 
constant, those increments are equal; if variable, they are un 

In 6, 7, and 8, the same statements may be made in regard to 
decrements. It is also plain that in these three last cases, if the 
force continues to act indefinitely, the motion will be retarded 
until the body comes to a state of momentary rest, and then is 
accelerated in the direction of the force. 

12. Measure of Force. The intensity of an impulsive force 
is measured by the momentum which it will produce or destroy ; 
that is, /oc m. But m oc q v\ :. fee q v. Hence, if q is con 
stant, / oc v. If, then, an impulse is applied to a given mass, the 
intensity of that impulse is measured by the velocity which it im 
parts or destroys. 

But in the case of a constant force, the momentum depends not 
only on the intensity of the force, but on the time during which it 

is applied; that IB, ft oc m, and/ oc . If the mass of the body 


is given, then, as in the case of an impulsive force, q being con- 


stant,/^ x v, and/oc -. 

To express the measure of a variable force, let t be a constant 
and infinitely small portion of time ; then the force varies as the 
mass multiplied by the increment of velocity imparted in that 

13. The Three Laws of Motion. All the phenomena of 
motion in Mechanics and Astronomy are found to be in accord 
ance with three first principles, which Newton announced in his 
Principia, and which are to be regarded as forming the basis of 
mechanical science. They may be named and defined as follows : 

1. The law of inertia. A body at rest tends to remain at rest ; 
and a body in motion tends to move forever, in a straight line, and 


2. The law of the coexistence of motions. If several motions 
are communicated to a body, it will ultimately be in the same 
position, whether those motions are simultaneous or successive. 

3. The law of action and reaction. If any kind of action takes 
place between two bodies, it produces equal momenta in opposite 
directions ; or, every action is accompanied by an equal and oppo 
site reaction. 

The truth of these laws cannot be established, except approxi 
mately, by direct experiments, because gravity, friction, and the 
resistance of air, interfere more or less with every possible experi 
ment. They are to be learned rather by a careful study of the 
phenomena of motion in general. We see an approximation to the 
first law, in rolling a ball on a horizontal surface ; first, on the 
earth, then on a floor, and again on smooth ice, the motion ap 
proaching toward uniformity as obstructions are diminished, and 
gravity producing no direct effect, because acting at right angles 
to the line of motion. The discussion of the second law is reserved 
for Chapter III. The third law is illustrated by a variety of cases 
in collision, attraction, and repulsion. Suppose that a body A, 
being in motion, strikes directly against B, which is at rest ; it is 
found that B acquires a certain momentum, and that A loses (that 
is, acquires in an opposite direction) an equal amount. The same 
is true if B is in motion, and A either overtakes or meets it. In 
the collision of two railroad trains, it is immaterial as to the 
effects which they will respectively suffer, whether each is moving 
towards the other, or whether one is at rest, provided that in the 
latter case the moving train has a momentum equal to the mo 
menta of the two trains in the former case. When a magnet 
attracts a piece of iron, each moves towards the other with the 
same momentum. A spring between two bodies A and B drives 
A from B with as much momentum as B from A ; and the sudden 
expansion of burning gunpowder, which propels the balls when a 
broadside is fired, causes an equal amount of motion of the ship in 
the opposite direction. 

14. Force of Gravity. Every mass of matter near the 
earth, when free to move, pursues a straight line towards its 
centre. The force by which this motion is produced is called 
gravity ; either the gravity of the body or the gravity of the earth ; 
for the attraction is mutual and equal, in accordance with the 
third law of motion. It is easy to understand why a small mass 
should attract a large one, as much as the large mass attracts the 
small one. Let A consist of one atom of matter, and B, at any 
distance from it, consist of ten atoms. If it be admitted that A 
attracts one atom of B as much as that one atom attracts A, then 


the above conclusion follows. For A attracts cadi of the ten 
atoms of B as much as each of the same ten attracts A ; so that . 1 
exerts ten units of attraction on B, while 1> exerts ten units of 
attraction on A. The same reasoning ohviously applies to the 
earth in relation to the small bodies on its surface. 

15. Relation of Gravity and Mass. At the same dis 
tance from the centre of the earth, gravity varies as the mass. 
This is because it operates equally on every atom of a body ; hence 
the greater the number of atoms in a body, the greater in the same 
ratio is the attraction exerted upon it. That gravity varies as the 
mass is also proved from the observed fact, that in a vacuum it 
gives the same velocity, in the same time, to every mass, however 
great or small, and of whatever species of matter. For a constant 
force, acting for a given time, is measured by the momentum which 
it produces (Art. 12), and that momentum, if the velocity is the 
same, varies as the mass : therefore the force also varies as the 
mass to which it imparts the given velocity. 

If a body is not free to move, its tendency towards the earth 
causes pressure; and the measure of this pressure is called the 
weight of the body. Weight is usually employed as a measure of 
the mass in bodies. The foregoing relations are embodied in the 
following expressions : g oc q ; and w oc q. 

16. Relation of Gravity and Distance. At different dis 
tances from the earth, gravity varies inversely as the square of the 
distance from the centre. The demonstration of this proposition is 
reserved for astronomy, where it is shown by the movements of 
the bodies in the solar system that this law applies to them all. 

The moon is 60 times as far from the earth s centre as the dis 
tance from that centre to the surface : therefore the attraction of 
the earth upon the particles of the moon is 3600 times less than 
upon particles at the surface of the earth. At the height of 4000 
miles above the earth, gravity is four times less than at the surface. 
But the heights at which experiments are commonly made upon 
the weights of bodies bear so small a ratio to the radius of the 
earth, that this variation is commonly imperceptible. At the 
height of half a mile, the diminution does not amount to more 
than about y^^th part of the weight at the surface. For, let 
r = the radius of the earth = 4000 miles, nearly ; and let x be the 
height of the body, w its weight at the earth s surface, and w its 
weight at the height x. Then, 
w :w : 

w : iv- w : : r^ + Zrx + x* : 2rx + x* .-. ww -^ *\> (A). 
But when x is a small fraction of r, a a may be neglected, and 


the formula becomes w w = - ~ ....... (E). 

Let x be half a mile; then = ^oi^ 1 P art f tne whole 

weight ; or, a body would weigh so much less at the height of half 
a mile than at the surface of the earth. But if the height were as 
great as 100 miles above the earth, the loss should be calculated 
by formula (A), since the other would give a result too small by 
one per cent, or more, according to the height. 

What loss of weight would a body sustain by being elevated 500 
miles above the earth? Ans. |y, or more than I of its weight. 

The relation of gravity to distance is expressed by the formula 

g x -jf ; and as g cc q also, it varies as the product of the two ; 

that is, g cc J--; or gravity towards the earth varies as the mass of 

the lody directly, and as the square of the distance from the earth s 
centre inversely. 

17, Gravity within a Hollow Sphere. A particle situated 
within a spherical shell of uniform density, is equally attracted in 
all directions, and remains at rest. This is true, because, in every 
direction from the body, the mass varies at the same rate as the 
square of the distance, so that attraction increases for one reason, 
as much as it diminishes for the other ; which is proved as follows : 

Let the particle P (Fig. 1) be at any point 
within the spherical shell A B CD. Let two 
opposite cones of revolution, of very small 
angle, have their vertices at P, and suppose the 
figure to be a section through the centre of the 
sphere and the axis of the cones. Then A B 
and a 1} will be the major axes of the small 
ellipses, which are the bases of the cones, and 
which may be considered as plane figures. By 
geometry, A P : P B :: P I : P a-, and the angles at P being 
equal, the triangles are similar ; hence the angles B and a arc 
equal. Therefore, the bases of the cones are similar ellipses, being 
sections of similar cones, equally inclined to the sides. By similar 
triangles, A P* : P 1? : A B* : a b*. Let q and q represent the 
masses of the thin laminae which form the bases ; then, since sim 
ilar ellipses are to each other as the squares of their major axes, we 
have , A D3 q q 


But -r and -~~ represent the attractions of the bases respec- 


tively on the particle (Art. 16) ; and since these are equal, the 
particle is equally attracted by all the opposite parts of the spheri 
cal shell. 

18. Gravity -within a Solid Sphere. "Within a solid sphere 
of uniform density, weight varies directly as the distance from the 

Let a particle P (Fig. 2) be within the solid FIG. 2. 

sphere A D G\ and call its distance from the 
centre d. Now, by the preceding article the 
shell exterior to it, A D R, exerts no influence 
upon it, and it is attracted only by the sphere 
P R 8. Let q represent the quantity of this 

sphere ; then gravity varies as . But q x d 3 ; 

d 3 
.: g <x -^ oc d. Hence, in the earth (if it be supposed spherical 

and uniformly dense, though it is neither exactly), a body at the 
depth of 1000 miles weighs three-fourths as much as at the surface, 
and at 2000 miles it weighs half as much, while at the centre it 
weighs nothing. 

Comparing this proposition with Art. 16, we learn that just at 
the surface of the earth a body weighs more than at any other 
place without or within. Within, the weight diminishes nearly z$ 
the distance from the centre diminishes ; without, it diminishes as 
the square of the distance from the centre increases. 

At the surface of spheres having the same density, iveight varies 
as the radius of the sphere. Let r be the radius of the sphere, and 

a r 3 

q its mass ; then, since g oc -*-, in this case it varies as ? QC r. 

Therefore, if two planets have equal densities, the weight of bodies 
upon them is as their radii or their diameters. If a bull two feet 
in diameter has the same density as the earth, a particle of dust at 
its surface is attracted by it nearly 21 millions of times less than it 
is by the earth. 

19. Questions for Practice. 

1. How much weight would a rock that weighs ten tons 
(22,400 Ibs.) at the level of the sea, lose if elevated to the top of a 
mountain five miles high? Ans. 55.8952 Ibs. 

2. If the earth were a hollow sphere, and if, through a hole 
bored through the centre, a man were let down by a rope, would 
the force required to support him be increased or diminished as 
he descended through the solid crust, and where would it become 
equal to nothing ? 

3. How much would a 44-pound shot weigh at the centre of 


the earth ; how much at a point half way from the centre to the 
surface; and how much 100 miles below the surface? 

4. If a hole were bored through the centre of the earth, and a 
stone were dropped into it, in what manner would the stone move 
in its way to the centre and after it reached the centre ? 
_ 5. Suppose a 32-pound cannon-ball, fired with the velocity of 
2,000 feet per second, to have the fciime momentum as a battering- 
ram whose weight is 5760 pounds; find the velocity of the latter. 

Ans. 11.11 ft. per sec. 

6. Suppose light to have weight, and one grain of it moving at 

, the rate of 192,000 miles per second, to impinge directly against a 

/ #f~ ) mass of ice moving at the rate of 1.45 feet per second, and to stop 

it ; require^! the weight of the ice. 

5 , * Ans. 99877.832 Ibs., or nearly 44.J tons, reckoning 7000 gr. = 1 Ib. 

&l B* ^ r 7. If a ball of the same density with the earth, j^th of a mile 

^_ 23J-^J? in diameter, were to fall through its own diameter toward the 

/* ^ $w** earthjjvhat space would the earth move through to meet the ball, 

i. - /? o W*w***tne 3mmeter of the earth being taken at 8,000 miles ? 

Ans. B ooo<jJoooo<j inch, nearly. 

8. Two men are pulling a boat ashore by a rope, one at each 
end, A being in the boat and B on the shore ; how will the time 
of bringing the boat ashore compare with the time in which A 
would pull it ashore alone, were the other end of the rope fixed to 
an immovable post ? 

9. Suppose the rope to pass from A in one boat to B in 
another equal boat ; how fast will B s boat move ? will A s boat 
have the same velocity as when B was on the shore ? 



20. Uniform Motion represented Geometrically. When 
a body moves uniformly for a given time, the space described 
equals the time multiplied by the velocity 
(Art. 6). Therefore, if one side of a rect- A 
angle represents the time of motion, and 
an adjacent side the uniform velocity, the 
area of the rectangle will represent the 
space described in that time, because the 
area equals the product of two adjacent 
sides. Thus, let A B, B C, &c. (Fig. 3), D 
represent any equal portions of time, and 
A F, B G, &c., the uniform velocity; 



FIG. 4. 

then A G, B II, &c., may be used to represent the spaces de 
scribed, and the rectangle A L may represent the space passed 
over with the velocity A F, in the time A E. 

21. Velocity Increased at Finite Intervals. Suppose 
the body to receive equal impulses at the beginning of all the 
equal portions of time, A B, B C, &c. (Fig. 4). Then, A F being 
the velocity given by the first im 
pulse, G H, K L, &c., the increments A 

of velocity, will each be equal to A F; 
and B H, the velocity during the B 
second portion of time, equals 2 A F , 
CL, that of the third, equals 3 A F, &c. c 
Therefore, B G, C K, D M, &c., are 
as 3, 2, 3, etc. But A B, A C, A D, 
&c., are as 1, 2, 3, &c. Hence the tri 
angles A B G, A C K, &c., are simi 
lar, and the same straight line A 
passes through the angles of all the rectangles. Now these rect 
angles represent the successive spaces described in the equal 
times, and their sum represents the whole space described in the 
time A E. This exceeds the triangle A E by the sum of the 
small equal triangles A F G, G H K, &c. 

22. Uniformly Accelerated Motion Represented. Let 

the increments of time and velocity (Fig. 5) be half as great as 
before. Then the sum of all the rect 
angles, or the whole space described, 
exceeds the triangle A E by the 
sum of the triangles Afg,gFG, &c. 
These triangles are one-half the sum 
of those in Fig. 4. Therefore, by con 
tinually halving the increments of 
time and velocity, the sum of the rect 
angles continually approaches the 
area of the triangle ; and when these 
increments become infinitely small, 
the first velocity becomes zero, and the sum of the rectangles 
equals the triangle. Therefore, the space described by a body 
which begins to move from rest by the action of a constant force 
may be represented by a right-angled triangle, as A E, whose 
side A E represents the time, and the side E the last acquired 

Such motion is said to be uniformly accelerated (Art. 11). An 
example of this is found in the fall of a body in a vacuum. For 
gravity acts incessantly, and within the range of our experiments 

FIG. 5. 















it may be considered as acting with equal intensity. The proper 
ties of the triangle enable us to ascertain very readily the laws of 
the fall of a body. 

23. Laws of the Fall of Bodies. When bodies fall from 
rest by the force of gravity, and unobstructed by the air, the fol 
lowing relations exist between the space, time, and velocity: 

1. The spaces vary as the squares of the times. 

2. The spaces vary as the squares of the acquired velocities. 

3. The times vary as the acquired velocities. 

For let s be the space described, v the velocity acquired by a 
body falling from rest for the time t, s r the space described, v the 
velocity acquired at any other period t of its fall ; then, from what 
has already been demonstrated, if t and t be represented by the 
lines A B and A D (Fig. 6), and v and v by the lines B C and 
D E, drawn at right angles to them, s and s will be FIG. 6. 
represented by the triangles A B C, A D E. Now, A. 

hence, s : s : : f : # 2 , or as v* : v f \ 

As equal increments of velocity are generated in 
equal times, it is farther evident that the velocity B 
acquired varies as the time ; the same conclusion 
may also be deduced from the similar triangles 

t: t . 

A B C, A DE- 7 for G\D E\\A B \ A D, i.e. v 

Since the spaces described are as the squares of the times, if 
a body falls from rest during times which are represented by the 
numbers 1, 2, 3, 4, 5, &c., the spaces described in those times will 
be as the square numbers 1, 4, 9, 16, 25, &c. ; and the spaces de 
scribed in equal successive portions of time will be as the 4 odd 
numbers 1, 3, 5, 7, 9, &c., as exhibited in the following table : 


Spaces described. 

Spaces described in equal successive portions of time. 







In ist port on of time 
3 d 


. . . 4- 1=3 
... 9- 4=5 
. . . 16 - 9 = 7 
. . . 25 16 = 9 
. . . &c. &c. 

The odd number expressing the space described in any unit 
of time, and which is found in the above table by taking the differ 
ence of the squares of successive numbers, may also be obtained by 
subtracting one from twice the number of units in the time. 
Thus, in the table, 3 = 2x2 1; 5 = 2x3 1, &c. This 
is true to any extent. Let n represent any whole number ; the 


number next less is n 1. The space described in n units of 
time is represented by w 2 , and that described in n 1 unii 
(n I) 3 . Therefore, the space described in the nth unit is ivp, - 
sented by ri* (n 1) : = 2 n 1. This is an odd number, 
and it equals twice the given number, less one. 

24. Uniformly Retarded Motion. If a body be projected 
perpendicularly upward in a vacuum, with the velocity which it 
has acquired in falling from any height, it will rise to the point 
from which it fell, before it begins to descend again, and the mo 
tion will be uniformly retarded. As the force of gravity adds 
equal velocities in equal times to a descending body, so it destroys 
equal velocities in equal times in a body which is ascending. The 
spaces described in successive units of time, by a body thus ascend 
ing, reckoning from the beginning of its motion, will be the same 
as those stated in the foregoing table, but in an inverted order : 
thus, if the time be divided into four equal parts, then the spaces 
described in the descent of the body during these equal times are 
as the numbers 1, 3, 5, 7, but in its ascent they will be as 7, 5, 3, 1 ; 
that is, the space described in the first portion of time, in its 
ascent, will be the same as that described in the last, in its de 
scent, and so on till the body arrives at its highest point. 

25. Acquired Velocity. If a body moves uniformly with 
the acquired velocity, it will pass over twice as great a space, in 
the same time, as it falls through to acquire it. 

Let the triangle A B C (Fig. 7) represent the A \ 
space described by gravity in the time A B, and 
B C the last acquired velocity ; produce A B to 
D, making B D equal to A B, and complete the 
rectangle B E\ then, if a body moves during the 
time B D with the uniform velocity represented 
by B C, the space described in that time will be 
represented by the rectangle B E\ but the tri 
angle A B C is half B E\ hence the space de 
scribed with the velocity B C continued uniformly 
is twice that which would be described in the same time A /> , 
falling from rest. 

Since the space described by a body falling from rest is half 
that which it would describe in the same time with its greatest 
velocity continued uniformly, and since a body projected per 
pendicularly upward rises to the same height as that from which 
it must fall to acquire the velocity of projection, the whole space 
described by a body projected perpendicularly upward is half that 
which it would describe in the same time with its first velocity 
continued uniformly. 



FIG. 8. 


26. Projection Downward. The space described in any 
time by a body projected doivnward with a given velocity is equal 
to the space which would be described with that velocity continued 
uniformly during that time, together with the space through which 
a body would fall from rest by the action of gravity in the same time. 

Let A D (Fig. 8) represent the given velocity of projection, and 
A B the given time, and complete the rectangle 
A E\ produce B E to (7, and let E G represent 
the velocity generated by gravity in the time A B 
or D E, and join D C. Then the body, moving 
by projection alone (Art. 20), would describe the 
rectangle A E in the time A B\ but, by gravity 
alone, it would describe the triangle DEC (Art. 
22). Hence, by the coexistence of both motions 
(Art. 13), it would describe the trapezoid A C. 

27. Projection Upward. The space described by a body 
ascending for a given time is equal to the space described uni 
formly with the velocity of projection in that time, diminished by 
the space fallen through from rest in the same time. 

Let B C (Fig. 9) be the velocity of projection, and A B the 
time in which a body would acquire that velocity FlG 9 
in falling from rest. Then the triangle ABC repre 
sents the space through which it would ascend before 
the velocity is lost. Let B E be the given time of 
ascent; then the rectangle B D is the space de 
scribed in the time B E, with the velocity B C con 
tinued uniformly, and CDF (similar to A B C) the 

space fallen through in the same time. But the part is" c 

B E F C of the triangle A B C is the space through which the 
body ascends in the time B E\ and this is equal to the difference 
of the rectangle B D and the triangle CDF. 

28. Formulae for the Fall of Bodies. The distance 
through which a body falls in a vacuum in one second of time 
varies on different parts of the earth. Between latitudes 40 and 
50, it is very nearly 16-^ feet, or 193 inches. Therefore (Art. 25), 
at the end of the second the body is moving with a velocity which, 
if gravity were to cease, would carry it over 32 J feet per second. 
Let g 32 \ feet, the velocity acquired in one second of fall. Then 
g 16 T 1 ^, the distance of fall in the first second. Let s be the 
space described, and v the velocity acquired, in any other time t. 
Then, according to the laws of variation (Art. 23), we have : 










FIG. 10. 

29. Space and Time Represented* by Co-ordinates. 

The relation of space and time in different kinds of motion may 
be well represented by the rectangular co-ordinates of certain lines. 
Thus, in uniform motion we have 

s = v t, 

in which v is constant. This may be 
regarded as the equation of a straight 
line passing through the origin, and 
making with the axis of abscissas an 
angle, whose tangent is v. Therefore, 
if any abscissa A G (Fig. 10) repre 
sents the number of units of time oc 

cupied in the motion, the corresponding ordinate C D will repre 
sent the space passed over. 

Again, for the uniformly accelerated motion of a falling body 

we have 


s = % g f , or F = - s. 


This is the equation of a parabola whose parameter is -. 
Therefore, if the parabola A B (Fig. 11) be described, having 


FIG. 11. 

- for its parameter, and the time of fall- 


ing is represented by any ordinate C D, 
the corresponding abscissa A C will 
represent the space fallen through. 

This is illustrated by Morin s appa 
ratus, where a body falls parallel to the 
axis of a uniformly revolving cylinder, 
wrapped with paper, against which a 
pencil, attached to the falling body, 
gently presses. When the paper is un 

wound and developed upon a plane, the curve traced by the pencil 
is found to be a parabola. 

30. Applications of Formulae for the Fall of Bodies.- 

1. A body falls G seconds ; what space does it pass over, and 
what velocity does it acquire ? Ans. a= 579 ft. v= 193 ft. per sec. 


2. How far must a body fall to acquire a velocity of 50 feet per 
second, and how long will it be in falling ? 

Ans. s = 38.86 ft. t = 1.55 sec. 

3. A body fell from the top of a tower 150 feet high ; hoiv long 
was it in falling, and what velocity did it have at the bottom ? 

Ans. t = 3.054 sec. v = 98.237 ft. 

4. If a ball be thrown upward with a velocity of 100 feet per 
second, what height will it reach ? Ans. 155.44 ft. 

5. Suppose a body to fall during 3 seconds, and then to move 
.uniformly during 2 seconds more, with the velocity acquired; 
what is the whole distance passed over? 

The space fallen through is 16 T 2 x 9 = 144| feet. The velo 
city acquired is 32^ x 3 = 96J feet. The space described uni 
formly is 96 J- x 2 = 193 feet. Therefore the whole space is 
144| + 193 = 337| feet. 

6. A ball fired perpendicularly upward was gone 10 seconds, 
when it returned to the same place ; how high did it rise, and with 
what velocity was it projected? Ans. s=402 T 1 3 ft., v=160g ft. 

31. Space in any Given Second or Seconds of Fall- 
Since the spaces described in the successive units of time are as 
the odd numbers, and as -J- g is described in the first second, there 
fore 3 x -|- g is described in the second, 5 x J g in the third, and 
generally (2^ 1) x \g in the nth second. 

1. How far does a body move in the 14th second of its fall? 

Ans. 434| ft. 

2. A body had been falling 2 minutes ; how far did it move in 
the last second? Ans. 3843 f* feet. 

The space described in the last m seconds is found thus : The 
space in the whole time t, = \ g f\ and in the time t m, the 
space 2 g (t my. Subtracting the latter from the former, we 
find the space described in the last m seconds to be ^g (2 m t m~). 
When m = 1, this becomes for the space in the last second A g 
(2 t 1). This is the same form of expression as was found 
above, where n was a whole number of seconds. Therefore, the 
space described in any one second of the fall, whether the time 
from the beginning is an integral or a fractional number, is found 
by multiplying J g by twice the number of seconds minus one. 

3. What space was described in the last two seconds by a body 
which had fallen 300 feet? Ans. 213.58 feet. 

4. A body had been falling 8A seconds ; how far did it descend 
in the next second ? Ans. 289 ft. 

32. Calculation for Projection Upward or Downward. 

A body projected downward describes t v feet by the force of pro 
jection, and 2 g t* feet by the force of gravity (Art, 26). A body 


projected upward describes t v by the force of projection ; but this 
is diminished by A g f, which gravity would cause it to describe in 
the same time (Art 27). Therefore the formula for space de 
scribed by a body projected downward is t v + \ g f ; by a body 
projected upward, the formula is t v % g t\ 

1. A body is projected downward with a velocity of 30 feet in a 
second; hoiu far will it fall in 4 seconds? Ans. 377J ft 

2. A body is projected upward with a velocity of 120 feet in a 
second; how far will it rise in 3 seconds? Ans. 215 J ft 

3. Suppose at the same instant that a body begins to fall from 
rest from the point D (Fig. 12), another body is projected 
upward from B with a velocity which would carry it to A ; FIG. 12. 
it is required to find the point where they would meet 

Let C be the point where the bodies would meet ; and 
let A B = , B D = b, D C x\ then will A D = a b, 
A C = a b + x. j 

/2 r\ - 
Now the time of descending through D C = I I j an( i 

the time of ascending through B C (= time down A B 
time down A C) = ( Y~~ ( ~^~> but the time 

down D C must be equal to the time up B C; hence we have - C 


9 ^9 x 9 
...( a _ i +z)i=0-i_a;3, 

.-. 2 (a x)% = b, or 4 a x = ~b\ and x = --. " B 

4 a 

4. Suppose a body to have fallen from A to B (Fig. 13), when 
another body begins to fall from rest at D\ how far will the latter 
body fall before it is overtaken by the former ? 

Let C be the point where one body overtakes the other, FIG. 13. 

Now time down D C = I Y , and time down B C time 

= I Y, 

i A n v J A -n ft ( a + % + 

down ^4 (7 time downed B ={ : 

V 9 

but at the moment when the lower body is overtaken, time 
down D C = time down B C, or 
/2 x\\_ /2 (a + b + 

l^/ "I 

33. Questions on Falling Bodies. 

1. The momentum of a meteoric stone at the instant of 



\f " ~ <S " ^ 7 I i- 

~ t 5 striking the earth was estimated at 18435 foot-pounds, and it had 


- t- been falling 10 seconds; from what height did it fall, and what 
1*3^ was its weight? Ans. 1608 J ft; 57.31 Ibs. 

2. An archer wishing to know the height of a tower, found 
that an arrow sent to the top of it occupied 8 seconds in going and 
returning; what was the height of the tower? Ans. 257^ ft. 

3. In what time would a man fall from a balloon three miles 
high, and what velocity would he acquire ? 

Ans. t = 31.38 sec. ; v = 1009.39 ft. 

4. A body having fallen for 3-J seconds, was afterwards observed 
to move with the velocity which it had acquired for 24 seconds 
more; what was the whole space described by the body? 

Ans. 478 \ ft., very nearly. 

5. Through what space would the aeronaut ^(in Question 3) fall 
during the last second ? Ans. 993.3 feet. 

6. A body has fallen from the top of a tower 340 feet high ; 
what was the space described by it in the last three seconds ? 

Ans. 298.957 ft. 

7. Suppose a body be projected downward with a velocity of 18 
feet in a second; how far will it descend in 15 seconds ? 

Ans. 3888| ft. 

8. A body is projected upward with a velocity of 65 feet in a 
second ; how far will it rise in two seconds ? A.ns. 65| ft. 

9. With what velocity must a stone be projected into a well 
450 feet deep, that it may arrive at the bottom in four seconds? 

Ans. 48 ftl in a second. 

10. The space described in the fourth second of fall was to the 
space described in the last second except four, as 1 : 3 ; what was 
the whole space described by the body ? Ans. 3618| ft. 

11. A staging is at the height of 84 feet above the earth. A 
ball thrown upward from the earth, after an absence of 7 seconds, 
fell on the staging ; what was the velocity of projection ? 

Ans. 124.58 ft. per second. 

12. A body is projected upward with a velocity of 483 feet in a 
second; in what time will it rise to a height of 1610 feet? 

Ans. t = 3.82 sec,, or 26.2 sec. 

13. From a point 214| feet above the earth a body is projected 
upward with a velocity of 161 feet in a second ; in what time will 
it reach the surface of the earth, and with what velocity will it 
strike ? Ans. t -- 11.2 sec., v 199 ft. 

14. Suppose a body to have fallen through 50 feet, when a 
second body begins to fall just 100 feet below it ; how far will 
the latter body fall before it is overtaken by the former ? 

Ans. 50 ft. 

15. A body is projected upward with a velocity of 64* feet in a 


FIG. 14. 

second; how far above the point of projection will it be at the 
of 4 seconds ? Am. ft. 

16. A body is projected upward with a velocity of 128f feet in 
a second; where will it be at the end of 10 seconds ? 

Ans. 321 ~ ft. below the point of projection. 

[See Appendix for* the discussion of the fall of bodies by the 

34. Atwood s Machine. Accurate observations on the di 
rect fall of a body cannot be easily 
made, on account of its great velocity ; 
and if they could be, the relations be 
tween time, space, and acquired velo 
city, would not be found to agree with 
those obtained by calculation, on ac 
count of the resistance of the air. Ex 
periments on falling bodies are usually 
performed by an instrument known as 
Atwood s machine, represented in Fig. 
14. From the base of the instrument, 
which is furnished with leveling screws, 
rises a substantial pillar, about seven 
feet high, supporting a small table upon 
the top. 

Above the table is a grooved wheel, 
delicately suspended on friction-wheels, 
and protected from dust by a glass case. 
Two equal poises, M and M 9 are at 
tached to the ends of a fine cord, which 
passes over the groove of the wheel. As 
gravity exerts equal forces on M and 
M , they are in equilibrium. To set 
them in motion, a small bar m is placed 
on M, which will immediately begin to 
descend, and M to rise. But this mo 
tion will be slower than in falling freely, 
because the force which gravity exerts 
on the bar must be communicated to 
the poises, and also to the revolving 
wheel over which the cord passes. By 
increasing the poises M, M , and dimin 
ishing the bar m, the motion may be 
made as slow as we please. // is a 
simple clock attached to the pillar for 
measuring seconds, and for dropping 
the poise M at the beginning of a vibra- 


tion of the pendulum. Q is a scale of inches extending from the 
base to the table. The stage A may be clamped to any part of the 
scale, in order to stop the poise M in its descent, as represented at 
G. The ring B, which is large enough to allow the poise, but not 
the bar, to pass through it, is also clamped to the scale wherever 
the acceleration is to cease. 

Let M be raised to the top, and held in place by a support, and 
then let the pendulum be set vibrating. "When the index passes 
the zero point, the clock causes the support to drop away, and the 
poise descends. The pendulum shows how many seconds elapse 
before the bar is arrested by the ring, and how many more before 
the poise strikes the stage. From the top to the ring the motion 
is accelerated by the constant fraction of gravity acting on it ; 
from the ring to the stage the poise moves uniformly with the 
acquired velocity. All the formulae relating to the fall of a body 
can therefore be illustrated by these slow motions. Moreover, the 
resistance of the air is so much diminished when the motion is 
slow, that a good degree of correspondence is found to exist be 
tween the experiments and the results of calculation. 

35. Living Force. We have seen that when a body is pro 
jected upward in a vacuum, the height to which it will rise varies 
as the square of the velocity of projection. In the ascent, the 
body is constantly and uniformly opposed by gravity. If the 
motion of a given body were opposed by any other uniform ob 
struction, the distance it would proceed before coming to rest 
would also vary as the square of the velocity. This power to over 
come a constant resistance varies, therefore, not as the mo 
mentum that is, as the product of mass and velocity q v but as 
the product of mass and square of velocity q v 2 ; and it is called 
the vis viva or living force, in distinction from vis inertia or dead 
force. If a ball weighing one pound move with the velocity of 
2,000 feet, and another ball weighing two pounds move with the 
velocity of 1,000 feet, then the momentum (q v) of the first equals 
that of the second. But the living force (q r 2 ) of the first is twice 
as great as that of the second ; for 1 x 2000 2 : 2 x 1000 2 : : 2 : 1. 
The first body, in its ascent, will reach four times as great a height 
as the second ; or, if the two balls be fired into a bank of earth, 
the first will penetrate four times as far as the second. 

In practical mechanics, the living force is generally called the 
working power ; and the luorJc which it will perform is therefore 
measured by the mass multiplied by the square of the velocity. 
The working power of the steam in a locomotive is employed in 
maintaining a certain velocity in the train, in spite of grade, fric 
tion of rails and machinery, and resistance of the air. If the power 


of the steam were wholly cut off, the train would be uniformly re 
tarded by the constant resistances, until, after running a certain 
distance, it would come to rest. But if the velocity of the train 
had been twice as great, it would have run four times as far before 
stopping ; it would also have required four times as great a force 
to give the train this double velocity. Both of these facts, one re 
lating to the effect, the other to the cause, show that the working 
power is to be estimated according to the square of the velocity. 
So the working power employed in moving any kind of machinery, 
which presents a constant resistance, varies as the square of the 
velocity imparted, and the work performed by the machinery is 
reckoned in the same way. To give a missile greater velocity is 
more advantageous than to increase its mass. A 40-pound ball, 
with 1400 feet velocity, is 7 times more efficient in penetrating 
the walls of forts and the hulks of ships than a 280-pound ball with 
200 feet velocity, though the momentum is the same in each case. 

36. Measure of Force. In Art. 12 force is said to be 
measured by momentum, orfccqv; and in Art. 35 it is said to be 
measured by the work performed, or / cc q v*. But these state 
ments are not to be considered as inconsistent with each other ; 
for in the first case, force has reference to inertia ; in the second 
case it has reference to work. When a force acts on a body that is 
free to move without obstruction (which is, however, only a sup- 
posable case), the effect is perpetual; the body will move on 
uniformly forever. If "the force had been greater, the velocity 
would have been greater in the same ratio. But when resistances 
oppose (as is always true in practice), then the force is expended 
in overcoming them, and this is the work to be performed ; and if 
the force ceases to operate, the motion will at length cease also ; 
but, as has been shown, the space passed over, and therefore the 
work performed, will vary as the square of the velocity. 

When force is employed to perform work, it is by some writers 
called energy, to distinguish it from force as used in producing 
momentum. Hence energy varies as the product of the mass and 
the square of the velocity. 



37. Motion by Two or More Forces. Motion produced 
by a single force, either impulsive or continued, has been already 
considered. But motion is more generally caused by several forces 
acting in different directions. 


"When two or more forces act at once on a body, each force is 
called a component, and the joint effect is called the resultant. 
Forces may be represented by the straight lines along which they 
would move a body in a given time ; the lines represent the forces 
in two particulars, the directions in which they act and their rela 
tive magnitudes. Whenever an arrow-head is placed on a line, it 
shows in which of the two directions along that line the force acts. 

38. The Parallelogram of Forces. This is the name 
given to the relation which exists betweeen any two components 
and their resultant, and is stated as follows : 

If two forces acting at once on a lody are represented by the 
adjacent sides of a parallelogram, their resultant is expressed ly the 
diagonal which passes through the intersection of those sides. 

Suppose that a body situated at A (Fig. 15) receives an impulse 
which, acting alone, would carry it 
over A B in a given time, and an 
other which would carry it over A 
in the same length of time. If both 
impulses are given at the same in 
stant, the body describes A D in the 
same time as A B by the first force, 
or A C by the second, and the motion in A D is uniform. 

This is an instance of the coexistence of motions, stated in the 
second law of motion (Art 13). For the body, in passing directly 
from A to D, is making progress in the direction A C as rapidly 
as though the force A B did not exist ; and at the same time it 
advances in the direction A B as fast as though that were the only 
force. When the body reaches D, it is as far from the line A B as 
if it had passed over A C; it is also as far from the line A (7 as if 
it had gone over A B. Thus it appears that both motions A B 
and A C fully coexist in the progress of the body along the diag 
onal A D. That the motion is uniform in the diagonal is evi 
dent from the law of inertia ; for the body is not acted on after it 
leaves A. 

It is evident that a single force might produce the same effect ; 
that force would be represented, both in direction and magnitude, 
by the line A D. The force A D is said to be equivalent to the 
two forces A B and A C. 

39. Velocities Represented. The lines A B and A C are 
described by the components separately, and the line A D by their 
joint action, in the same length of time. Hence the velocities in 
those lines are as the lines themselves. In the parallelogram of 
forces, therefore, two adjacent sides and the diagonal between them 


1st The directions of the components and resultant ; 

2d. Their relative magnitudes ; and 

3d. The relative velocities with which the lines are described. 

40. The Triangle of Forces. For purposes of calculation, 
it is more convenient to represent two components and their re 
sultant by the sides of a triangle, than by the sides and diagonal 
of a parallelogram. In Fig. 15, C D, which is equal and parallel to 
A B, may represent in direction and magnitude the same force 
which A B represents. Therefore, the components are A and 
CD } while their resultant is A D; and the angle Cm the triangle 
is the supplement of A B, the angle between the components. 
Care should be taken to* construct the triangle so that the sides 
representing the components may be taken in succession in the 
directions of the forces, as, A C, C D ; then A D correctly repre 
sents their resultant. But, although A C and A B represent the 
components, the third side, C B, of the triangle A C B, does not 
represent their resultant, since A C and A B cannot be taken suc 
cessively in the direction of the forces. It is necessary to go back 
to A in order to trace the line A B. It should be observed, that 
though C D represents the magnitude and direction of the compo 
nent, it is not in the line of its action, because both forces act 
through the same point A. 

41. The Forces Represented Trigonometrically.- Since 

the sides of a triangle are proportional to the sines of the opposite 
angles, these sines may also represent two components and their 
resultant. Thus, the sine ofCAD corresponds to the component 
A B (= CD); the sine of CD A (= D A B) corresponds to the 
component A C; and the sine of C (= sine of C A B) corresponds 
to the resultant A D. Each of the three forces is therefore repre 
sented by the sine of the angle between the other two. 

42. Greatest and Least Values of the Resultant A 

change in the angle between the components alters the value of 
the resultant; as the angle increases from to 180, the resultant 
diminishes from the sum of 
the components to their differ- Fia 

ence. In Fig. 16, let C A B 
and D A B be two different 
angles between the same com 
ponents A C (or A D) and 
A B. As C A B is less than 
DAB, its supplement A B F 
is greater than A B E, the 
supplement of DA B; there 
fore A F is greater than A E. When the angle C A B is, dimin- 


FIG. 17. 

ished to 0, the sides A B, B F, become one straight line, and A F 
equals their sum; when D A B is enlarged to 180, E falls on 
A B, and A E equals the difference of A B and A C. Between 
the sum and difference of the components, the resultant may have 
all possible values. 

43. The Polygon of Forces. All the sides of a polygon 
except one may represent so many forces acting at the same time 
on a body, and the remaining side will represent their resultant. 
In Fig. 17, suppose A B, A 0, 
and A D, to represent three 
forces acting together on a body 
at A. The resultant of A B and 
A C is represented by the diag 
onal A E-y and the resultant 
of AE and A D by the diagonal 
A F. As A F is equivalent to 
A E and A D, and A E is equiv 
alent to A B and A C, therefore 
A F is equivalent to the three, A B, A C, and A D. But if we 
substitute B E for A C, and E F for A D, then the three compo 
nents are A B, B E, and E F, three sides of a polygon, and the 
resultant A F is the fourth side of the same polygon. 

So, in Fig. 18, A B, B C, C D, D E, and E F, may represent 
the directions and relative magni 
tudes of five forces, which act simul 
taneously on a body at A. The re 
sultant of A B and B C is A (7; the 
resultant of A C and C D is A D ; 
the resultant of A D and D E is 
A E-, and the resultant of A E and 
JZFis A F-, which last is therefore 
the resultant of all the forces, A B, 
B C, CD, DE, and^^ 7 ; the com 
ponents being represented by five 
sides, and their resultant by the sixth side, of a polygon of six 

FIG. 18. 


44. Curvilinear Motion. Since, according to the first law 
of motion, a moving body proceeds in a straight line, if no force 
disturbs it, whenever we find a body describing a curve, it is cer 
tain that some force is continually deflecting it from a straight 
line. Besides the original impulse, .therefore, which gave it 
motion in one direction, it is subject to the action of a continued 
force, which operates in another direction. A familiar example 
occurs in the path of a projectile. Suppose a body to be thrown 



FIG. 19. 

from P (Fig. 19), with an impulse which would alone carry it to 

N, in the same time in which gravity alone would carry it to V. 

Complete the parallelogram P Q; 

then, as both motions coexist (2d 

law), the body at the end of the 

time will be found at Q. Let t be 

the time of describing P Nor P V\ 

and let t be the time of describing 

P M by the impulse, or P L by 

gravity. Then, at the end of the 

time t , the body will be at 0. Now, 

as P N is described uniformly, 


But (Art M),Pr:PL::f: t n ; 

.-. P V: P L : : P N* : P M*; or Q V* : L\ 

Hence, the curve is such that P Fee Q F a ; that is, the abscissa 
varies as the square of the ordinate, which is a property of the 
parabola. P Q is therefore a parabola, one of whose diameters 

is P F, and the parameter to that diameter is -p-y 

Owing to the resistance of the air, the curve deviates sensibly 
from a parabola, especially in swift motions. 

45. Calculation of the Resultant of Two Impulsive 
Forces. When two components and the angle between them are 
given, the resultant may be found both in direction and magni 
tude by trigonometry. The theorem required is that for solving a 
triangle, when two sides and the included angle are given ; but the 
included angle is not that between the components, but its supple 
ment (Art 40). In Fig. 15, if A B = 54, and A C = 22, and 
C A B = 75, then A C D is the triangle for solution, in which 
A C= 22, CD = 54, and A CD = 105. Performing the cal 
culation, we find the resultant A D = 63.363, and the angle/) A B, 
which it makes with the greater force, = 19 35 43". This 
method will apply in all cases. 

1. A foot-ball received two blows at the same instant, one directly 
east, at the rate of 71 feet per second, the other exactly northwest, 
at the rate of 48 feet per second ; in what direction and with what 
velocity did it move ? Ans. N. 47 30 52" E. Vel. = 50.253. 

The process is of course abridged, if the forces act at a right 
angle with each other, as in the following example : 

2. A balloon rises 1120 feet in one minute, and in the same 
time is borne by the wind 370 feet ; what angle does its path make 
with the vertical, and what is its velocity per second? 

Ans. 18 16 53"; v = 19.659. 


In the next example, one component and the angle which each 
component makes with the resultant, are given to find the result 
ant and the other component. 

3. From an island in the Straits of Sunda, we sailed S. E. by S. 
(33 45 ) at the rate of 6 miles an hour; and being carried by a 
current, which was running toward the S. "WV (making an angle 
with the meridian of 64 12 J ), at the end of four hours we came 
to anchor on the coast of Java, and found the said island bearing 
due north ; required the length of the line actually described by the 
ship, and the velocity of the current 9 

Ans. s = 26.4 miles. 

v 3.7024 miles per hour. 

If the magnitudes and directions of any number of forces are 
given, the resultant of them all is obtained by a repetition of the 
same process as for two. In Fig. 18, first calculate A C, and the 
angle A C B, by means of A B, B C, and the angle B. Subtract 
ing A C B from B C D, we have the same data in the next tri 
angle, to calculate A D, and thus proceed to the final resultant, 

As it is immaterial in what order the components are intro 
duced into the calculation, it will diminish labor, to find first 
the resultant of any two equal compo 
nents, or any two which make a right 
angle with each other ; since it can be 
done by the solution of an isosceles, or 
a right-angled triangle. 

4. The particle A (Fig. 20) is urged 
by three equal forces A B, A C, and 
A D-, the angle B A C = 90, and 

A D = 45 ; what is the direction of A 
the resultant, and how many times A B ? 
Ans.BAF = 80 16 , and 
A F : A B : : |/3 : 1. 

5. Five sailors raise a weight by 
means of five separate ropes, in the same 

plane, connected with the main rope that is fastened to the weight 
in the manner represented in Fig. 22. B pulls at an angle with 
A of 20 ; C with B, 19 ; D with (7, 21 30 ; and E with Z>, 25. 
A, B, and C, pull with equal forces, and D and E with forces one- 
half greater; required the magnitude and direction of the re 

Ans. Its angle with A is 46 33 10". Its magnitude is 5.1957 
times the force of A. 

If the polygon A B C D E (Fig. 21) be constructed for the 
above case, A C and D F are easily calculated in the isosceles tri- 

FIG. 20. 


angles A B C and D E F, after which A D and then A F are to 
be obtained by the general theorem. 


FIG. 23. 

FIG. 23. 

46. The Resultant and all Components, except one, 
being given, to Find that one Component. If A B (Fig. 
23) is the resultant to be produced, and 
there already exists the force A C, a, 
second force can be found, which acting 
jointly with A C, will produce the mo 
tion required. Join C B, and draw A D 
equal and parallel to it, then A D is the 
force required ; for A B is equivalent to 
A C and C B. Therefore C B has the 
magnitude and direction of the required 
force ; A D is the line in which it must 

Again, suppose that several forces act on A, and it is required 
to find the force which, in conjunction with them all, shall pro 
duce the resultant A B. Let the several forces be combined into 
one resultant, and let A C represent that resultant. Then A D 
may be found as before. 

The trigonometrical process for finding a component is essen 
tially the same as for finding a resultant. 

1. A ferry-boat crosses a river f of a mile broad in 45 minutes, 
the current running all the way at the rate of 3 miles an hour ; at 
what angle with the direct course must the boat head up the stream 
in order to move perpendicularly across? Ans. 71 34 . 


2. A sloop is bound from the mainland of Africa to an island 
bearing W. by N. (78 45 ) distant 76 miles, a current setting 
N. N". W. (22 30 ) 3 miles an hour ; what is the course to arrive 
at the island in the shortest time, supposing the sloop to sail at the 
rate of 6 Mws^er hour ; and what time will she take ? 

Ans. Course, S. 76 41 4" W. Time, 10 h. 40m. 7 sec. 

3. The resultant of two forces is 10 ; one of them is 8, and the 
direction of the other is inclined to the resultant at an angle of 36. 
Tind the angle between the two forces. 

Ans. 47 17 5" or 132 42 55". 

4. A ball receives two impulses : one of which would carry it 
N. 27 feet per second ; the other, E. 30 N. with the same velocity ; 
what third impulse must be conjoined with them, to make the ball 
go E. with a velocity of 21 feet ? Ans. S. 3 22 W. v = 40.57. 

47. Resolution of Motion. In the composition of motions 
or forces, the resultant of any given components is found ; in the 
resolution of motion or force, the process is reversed; the resultant 
being given, the components are found, which are equivalent to 
that resultant. 

If it be required to find what two components can produce the 
resultant A B (Fig. 24), we have 
only to construct on A B, as a base, 
any triangle whatever, as A B C or 
A B D (Art. 40); then, if A C is J> 
one component, the other is A F, 
equal and parallel to CB ; or if A D 
is one, the other is A E, equal and 
parallel to D B ; and so for any tri 
angle whatever on the base A B. 

F "******. J* 

The number of pairs is therefore in 
finite, whose resultant in each case is A B. 

The directions of the components may be chosen at pleasure, 
provided the sum of the angles made with A B is less than two 
right angles. 

The magnitude and direction of one component may be fixed 
at pleasure. 

The magnitudes of both components may be what we please, 
provided their difference is not greater, and their sum not less, 
than the given resultant. 

These conditions are-obvious from the properties of the triangle. 

When a given force has been resolved into two others, each of 
those may again be resolved into two, each of those into two others 
still, and so on. Hence it appears that a given force may be re 
solved into any number of components whatever, with such limi- 


tations as to direction and magnitude as accord with the foregoing 

1. A motion of 153 toward the north is produced by the forces 
100 and 125 ; how are they inclined to the meridian ? 

Ans. 54 28 and 40 37 7". 

2. A resultant of 617 divides the angle between its components 
into 28 and 74 ; what are the components ? 

Ans. 606.34 and 296.14. 

48. Resolution into Pairs, with Certain Conditions. 

In some cases, in which a condition is imposed, a simple construe* 
tion will enable us to find the pairs of components which fulfill 
that condition. 

1st. A given force is to be resolved into pairs which make a 
given angle with each other. 

Let A B (Fig. 25) be the given force. On A B as a chord, 
construct the segment of a circle A D JB, con 
taining an angle equal to the supplement of 
the given angle. Then all the possible pairs of 
components fulfilling the condition will be 
found by drawing lines from A and B to points 
of the curve, as A D, D B, and A C, C B, &c. 
The segment must contain, not the given angle itself, but its sup 
plement, because the given angle is at A, between A D and a 
parallel to D B, or C B, &c. 

If the given angle were a right angle, the segment to be con 
structed is a semicircle. 

2d. To resolve a given force into two components, making a 
given sum; the sum must not be less than the given force 
(Art. 47). 

Let A B (Fig. 26) be the given force, and N N the given 
sum. Having placed A B on M N, 
so that A N= B M, construct a semi- Fl - 2<3. 

ellipse on M N as a transverse axis, 
with A and B for the foci. Lines D 

drawn from A and B to any point of 

the curve will represent a pair of the 

required components ; for, by a prop- N A B M 

erty of the ellipse, A D + D B, or 

A C + CB = M N. 

3d. To resolve a given force into two. components, having a 
given difference ; the difference must not be greater than the given 
force (Art. 47). 

Let A B (Fig. 27) be the given force, and M N the given dif 
ference. Place M N on A B so that A N = B J/, and construct 


FIG. 27, 

the hyperbola M C, N D, having M JVfor its transverse axis, with 

A and B for foci. Then A D, D B, or A C, C B, or any other 

lines from the foci to a point 

of the curve, will fulfill the 

condition required, because 

their difference equals the 

transverse axis. 

It is required to resolve 
the force 194 into pairs of 
components acting at an an 
gle of 135 with each other ; 
what is the radius of the circle whose segment is employed in the 
construction ? Let A B 194 (Fig. 28), and A D B 45 ; 
then A C B = 90. Let Clfbe perpendicular 

A B FIG. 28. 

to A B ; then A B : A C :: A C : -^ , and 

A C = 

= 137.18. 

2. To find the radius of the circle whose seg 
ment includes the components of the force a 
acting at any given angle with each other. A 
Make A B = a, and let A D B = A, the sup 
plement of the given angle. Then A C B 2 A, and A C H 

2 sin A 

A ; 

sin A : 1 : : : A C 

49. Resolution of a Force, to Find its Efficiency in a 
Given Direction. By the resolution of a force into two others 
acting at right angles with each other, it is ascertained how much 
efficiency it exerts to produce motion in any given direction. For 
example, a weight W (Fig. 29), lying on a horizontal plane, and 

pulled by the oblique force C A, is prevented by gravity from 
moving in the line C A, and is compelled to remain on the plane. 
Resolve C A into C B, in the plane, and G D perpendicular to it ; 
then the former represents the component which is efficient to 



cause motion along the plane ; the latter has no influence to aid or 
hinder that motion ; it simply diminishes pressure upon the plane. 
In like manner, if A C is an oblique force, pushing the weight, its 
horizontal component, B C y is alone etlicient to move it; the other, 
A By merely increasing the pressure. In either case, the whole 
force is to that component which is efficient to move the body 
along the plane, as radius to the cosine of inclination. Also, the 
whole force is to that component which increases or diminishes 
pressure on the plane, as radius to the sine of inclination. 

If only 88 per cent, of the strength of a horse is efficient in 
moving a boat along a canal, what angle does the rope make with 
the line of the tow-path? Ans. 28 21 27". 

50. Resultant found by means of Rectangular Axes. 

When several forces act in one plane upon a body, their resultant 
may be conveniently found by the use of right-angled triangles 
alone. Select at pleasure two lines at right angles to each other, 
both of them lying in the plane of the forces, and passing through 
the point at which the forces are applied. These lines are called 
axes. The following example illustrates their use : 

Let P A, P B, P (7, P A P E (Fig. 30) represent the forces 
in Question 5 (Art. 45). Let one axis, for 
convenience, be chosen in the direction P A, 
and let P If be drawn at right angles to it 
for the other axis. These axes are supposed 
to be of indefinite length. Then proceed as 
in Art. 49, to resolve each force into two 
components on these axes. As P A acts in 
the direction of one axis, it does not need to 
be resolved. To resolve P B, say 

FIG. 30. 


R : cos 20 
R : sin 20 

P B : P I, and 
P B : PV 


: cos 39 
:sm 39 

: : P <7:Pe, and 
::P C:Pc ,&c. 

Suppose P A produced so as to equal PA 4- P ft + P c + 
p c i + p e M, and P # produced so as to equal P V + P c + 
P d + P e N. Now, as M acts in the line P A, and N at 
right angles to it, their resultant and the angle which it makes 
with P A are found by the solution of another right-angled tri 
angle. The resultant is 5.1957, and the angle is 46 33 10", as in 
Art. 45. 

If any components of the resolved forces are opposite to P A 
or P H, they are reckoned as negative quantities. 


FIG. 31. 

51. Analytical Expression for the Resultant, Put A C 
(Fig. 31) = P,A = P , AD = R, angle C A B = a; then by 

A D* = A C* + CD* - 2 A C x CD cos A CD, or 
R* = P 2 + P 2 + 2 P P cos a; whence 
P = 4/P* + P 2 + 2 P P 7 ^^ . . . (1). Hence 
Tlie resultant of any two forces, act 
ing at the same point, is equal to the 
square root of the sum of the squares 
of the two forces, plus twice the pro 
duct of the forces into the cosine of the 
included angle. 

If a = 0, its cosine will be 1, and 

(1) becomes 

R = P + P . 

If a = 90, its cosine will be 0, and we shall have 

R = |/P 2 + P \ 
If a = 180, its cosine will be 1, and we shall have . 

R = P - P . 

1. Two forces, P and P , are equal in intensity to 24 and 30, 
respectively, and the angle between them is 105 ; what is the in 
tensity of their resultant? Ans. 33.21. 

2. Two forces, P and P , whose intensities are, respectively, 
equal to 5 and 12, have a resultant whose intensity is 13 ; required 
the angle between them. Ans. 90. 

3. A boat is impelled by the current at the rate of 4 miles per 
hour, and by the wind at the rate of 7 miles per hour ; what will 
be her rate per hour when the direction of the wind makes an 
angle of 45 with that of the current ? Ans. 10.2 miles. 

4. Two forces and their resultant are all equal ; what is the 
value of the angle between the two forces ? Ans. 120. 

52. Principle of Moments. The moment of a force, with 
respect to a point, is the 

product of the force into 
the perpendicular let fall 
from the point to the 
line of direction of the 

The fixed point is 
called the centre of mo 
ments; the perpendicu 
lar distance, the lever- 
arm of the force; and 
the moment measures the tendency of the force to produce rotation 


about tlie centre of moments. Denote the two forces A D, A /!, 
and their resultant A C (Fig. 32) by P, P , and Jt, respectively. 
From E, any point in the plane of the forces, let fall, upon the 
directions of the forces, the perpendiculars E F, EH, E G. Repre 
sent these perpendiculars by p, p , r. Draw D K and B L perpen 
dicular to A C. Puta=CAD,(*=CAB,0 = CA E. Then 

R = A L + C L = P cos )3 + P cos a . . . (1) ; 
D K = P sin 0, B L = P sin a ; .-., since D K = B L, 
P sin j3 = P sin a, or, P sin P sin a . . . (2). 
Multiplying both members of (1) by sin 6, both members of (2) 
by cos 0, adding and reducing, we have 

R sin 6 = P sin (0 + (3) + P sm (0 a) . . . (3). 

But sin = -JL, sin (0 + 0) = -^, sin (0 - a) = -^; 

.-. (3) reduces to 

Rr = P p + Pp... (4). 

If the point E falls within the angle CAD, sin (0 a) be 
comes negative, and (3) becomes 

R r = P p -Pp... (5). 

Hence, the moment of the resultant of two forces is equal to the 
algebraic sum of the moments of the forces taken separately. 

53. Forces Acting at Different Points. Parallel 
Forces. We have thus far considered forces acting upon a 
single particle, or upon one 
point of a body. If, how 
ever, two forces P and P y 
in the same plane, act upon 
A and B, two different 
points of a rigid body, they 
may still have a resultant. 

Let the lines of direc 
tions of the two forces A F and E D (Fig. 33) be produced to 
meet in C. The two forces may then be considered as acting at 
C, and thus compounded into a single force at that point, or at 
the point G of the body. 

By (1) of the last article this resultant is 

R = P cos0 + Pcosa... (1). 

When the forces become parallel, as A F and B E, (3 = 0, and 
a = 0, and (1) becomes 

R = p + p . . . (2). 

If the parallel forces act in opposite directions, as A F and 
B E , then a = 180, and = 0, and (1) becomes 
R = p - p . . . (3). Hence, 


The resultant of tivo parallel forces is in a direction parallel to 
them and equal to their algebraic sum. 

54. Point of Application of the Resultant. Let P and 

P (Figs. 34, 35) be two parallel forces acting in the same or in 

FIG. 34. FIG. 35. 


opposite directions, and let E be the point of application of the 
resultant. Assume this point as a centre of moments ; then from 
(5) of Art. 52, since r 0, 

P x H E = P x G E, or, in the form of a proportion, 
P : P : : H E : G K But by similar triangles, 
HE-. QE-.\AE\EB\ :. 
P 1 : P : : A E : E B. 

That is, the line of direction of the resultant of two parallel forces 
divides the line joining the points of application of the components, 
inversely as the components. 

By composition (Fig. 34) and division (Fig. 35) we obtain - 
P + P : P : : A B : E B, and 
P - P 1 : P : : A B : E B. 

That is, if a straight line be drawn to meet the lines of two parallel 
forces and their resultant, each of the three forces will be propor 
tional to that part of the line contained between the other two. 
When the forces act in the same direction, we have 

P x A B 
E B = , , and when they act in opposite directions, 

If, in the last case, P P , then E B will be infinite. The 
two forces in this case constitute what is called a couple. Their 
effect is to produce rotation about a point between them. 

Any number of parallel forces may be reduced to a single force 
(or to a couple) by first finding the resultant of two forces, then 
the resultant of that and a third force, and so on to the last. And 
any single force may be resolved into two or any number of paral 
lel forces by a method the reverse of this. 

55. The Parallelepiped of Forces. Hitherto forces have 
been considered as acting in the same plane. But if forces act in 



FIG. 36. 

different planes, the solution of every case may be reduced to the 
following principle, called the parallelopiped of forces. 

Any three forces acting in different planes upon a body may be 
represented by the adjacent edges of a parallelopiped, and their re 
sultant by the diagonal which passes through the intersection of 
those edges. 

Let A C, A D, and A E (Fig. 36), be three forces applied in 
different planes to the body at A. 
Construct the parallelopiped G P, 
having A C, A D 9 and A E, for its 
adjacent edges, and from A draw the 
diagonal A B. The section through 
the opposite edges A C and P B is a 
parallelogram, and therefore A B is 
the resultant of A C and A P, and 
A P is the resultant of AD and A E. 
Hence A B is the resultant of A G, 
A D, and A E. 

This process may obviously be reversed, and a given force may 
be resolved into three components in different planes along the 
edges of a parallelopiped, having such inclinations as we please. 

56. Rectangular Axes. The parallelopiped generally 
chosen is that whose sides are rectangles; the three adjacent 
edges of such a solid are called rectangular axes. All the forces 
which can possibly act on a body may be resolved into equivalent 
forces in the direction of three such axes. And since all forces 
which act in the direction of any one line may be reduced to a 
single force by taking their algebraic sum, therefore any number 
of forces acting through one point may be reduced to three in the 
direction of three axes chosen at pleasure. 

Let A X, A Y (Fig. 37) be at right angles with each other, 
FIG. 37. FIG. 38. 

and A Z perpendicular to the plane of A X and A Y. Let A B 
represent a force acting on A. Eesolve A B into A C on the axis 



A Z, and A P in the plane of A X, A Y\ then resolve A P into 
A D and A E on the other two axes. Therefore, A C, A D, and 
A E are three rectangular forces, whose resultant is A B. 

Let the axes A X, AY,AZ, be produced indefinitely (Fig. 38) 
to X , Y , Z , then their planes will divide the angular space 
about A into eight solid right angles, namely : A-X YZ, A-X Y Z, 
A-X Y Z, A-X YZ, above the plane of X and Y, and A-X YZ , 
A-X Y Z , A-X Y Z , A-X YZ below it. 

FIG. 39. 

57. Geometrical Relation of Components and 
sultant. A force acting on the body 
A may be situated in any one of the 
eight angles, and its value may be 
expressed in terms of the squares of 
its three components. Let A B (Fig. 
39) be resolved as before into the 
rectangular components A C, A D, 
and A E. Then, by the right-an 
gled triangles, we find 



and A B = VA C* + A D* + A E\ 

If A B is in the plane of X and Y, the component on the axis 
of Z becomes zero, and A B = VA C* + A D*, and similarly for 
the other planes. 

58. Trigonometrical Relation of Components and Re 
sultant. Let the angles which A B makes with the axes of 
X, Y, Z, respectively, be a, ft y ; that is, B A C = a, B A D = (3, 
B A E = y. In the triangle ABC, right-angled at C, we have 
A B : A C : : rad : cos a therefore, making rad 1, 

A C = A B . cos a. 

In like manner, A D = A B . cos (3 ; 

and A E = A B . cos y. 

And since A B is the resultant of the forces A C, A D, and 
A E, it is the resultant of A B . cos a, A B . cos (3, A B . cos y. 
In general, the components of any force P, when resolved upon 
three rectangular axes, are P . cos a, P . cos (3, P . cos y. 

59. Any Number of Forces Reduced to Three on 
Three Rectangular Axes. Suppose the body at A to be acted 
upon by a second force P , whose direction makes with the axes 
the angles a, f3 9 y ; then, as before, P is the resultant of P . cos a , 
P .cosT? , .P. cos} ; and a third force P", in like manner, has 


for its components I" . cos a", I" . cos p", I" . cos / ; and so of 
any number of forces. 

Now, all the components on one axis may be reduced to one 
force by adding them together. Hence, the whole force in the 
axis of X - P . cos a + P . cos a + P" . cos a" + P" . cos a " -f &c. ; 
the whole in the axis of Y, 

= P . cos j3 + F . cos ft 1 + F" . cos ft" + P " . cos ft" + &c.; 
and that in the axis of Z, 

= P.cosy + P . cos / + T" . cos y" + P " . cos y " + &c. 

If any component acts in a direction opposite to others in the 
same axis, it is affected by a contrary sign, so that the force in the 
direction of any axis is the algebraic sum of all the individual 
forces in that axis. 

If the sum of the components in one axis is reduced to zero by 
contrary signs, the effect of all the forces is limited to the plane of 
the other axes, and is to be obtained as in Art. 50, where two axes 
were employed. If the sum of the components on each of two axes 
is reduced to zero, then the whole force is exerted in the direction 
of the remaining axis, and is therefore perpendicular to the plane 
of the other two. 

60. Equilibrium of Forces. 

1. Two forces produce equilibrium wlien they are equal and act 
in opposite directions. 

It was shown (Art. 42) that two forces produce the least re 
sultant when they act at an angle of 180 with each other, and 
that the resultant then equals the difference of the forces. If the 
forces are equal, their difference is zero, and the resultant vanishes ; 
that is, the two forces produce equilibrium. 

2. Three forces produce, equilibrium when they may be repre 
sented in direction and magnitude by the three sides of a triangle 
taken in order. 

For, when three forces are in equilibrium, one of them must 
be equal to, and opposite to, the re 
sultant of both the others. But ^ Q - 40 - 
the forces A C and A B (Fig. 40) 
produce the resultant A D; there 
fore the equal and opposite force 
D A, since it is in equilibrium with 
A D, is also in equilibrium with AC -A- -B 
and A B, or A C and C D. Hence 

the three forces A C\ CD, and D A, taken in order around the 
figure, produce equilibrium. 


It is obvious tbat three forces in equilibrium must all be di 
rected through one point, else each force could not be opposed to 
the resultant of the other two. 

3. More than three forces in one plane will produce equilibrium 
when they can be represented by the sides of a iwlygon taken in 

In Art. 43 it was shown that if several forces acting on a body, 
are represented by all the sides of a polygon .except one, their re 
sultant is represented by the remaining side. Thus, the resultant 
of the forces A B, B C, C D, and D E (Fig. 41), is A E. Now, 
the force E A, equal and opposite to A E, since it would be in 
equilibrium with A E, is therefore in equilibrium with all the 
others. Hence the forces A B, B C, C D, D E, and E A, taken 
in order around the figure, are in equilibrium. 

FIG. 41. FIG. 42. 

61. Trigonometrical Representation of Three Forces 
in Equilibrium. When three forces are in equilibrium, each may 
be represented by the sine of the angle between the other two. 

Let A E (Fig. 42) be the resultant of A D and A (7; then, if 
we apply a force A B equal and opposite to A E, the forces A D, 
A C, and A B will be in equilibrium. From the triangle A ED 
we have the proportions 

AD:AE\ ED ::sinA ED : sin D : sin E A D. 

But sin^^D sin CAE=s,mA (7; sinZ> sin CAD , 
sin E A D = sin B A D , A E = A B , and A C = ED; . . 
ADiAB-.A CusinBA <7:sin CA D: sin AD. 

62. Equilibrium of Parallel Forces. In order that a force 
may be in equilibrium with two parallel forces, 

1. It must be parallel to them. 

2. It must be equal to their algebraic sum. 

3. The distances of its line of action from the lines in which the 
two forces act, must be inversely as the forces. 

These three conditions belong to the resultant of two parallel 
forces, and therefore belong to that force which is in equilibrium 
with the resultant. 


63. Equilibrium of Couples. If two parallel forces uiv 
such as to constitute a couple, no one force can be in equilibrium 
with them. For the resultant of a couple has its point of applica 
tion at an infinite distance (Art. 54). But a couple can be held in 
equilibrium by another couple; and the second couple maybe 
either larger or smaller than the given couple, or it may be equal 
to it. 

Let the couple P and P (Fig. 43) act FIG. 43. 

on a body at the points A and B ; they 
tend to produce rotation about the middle 
point C. If another couple, Q and Q , 

equal to P and P\ should be applied to 

produce equilibrium, one must directly ~P 

oppose P, and the other P . Then A and 


B) being each held at rest, all the forces p-p 

are in equilibrium. 

But if the second couple is less than J? 

P and P , they must act at distances from 
(7, which are as much greater as the forces 
are less ; or, if the second couple is greater 
than the first, they must act at distances pl 

which are as much less. Thus, the couple 

p and p , acting at D and E, tend to produce rotation about C in 
one direction, and P and P in the opposite ; and these tendencies 
are equal when D C : A C : : P : p. For, since the opposite forces, 
P and p, are inversely as their distances from (7, their resultant is 
at (7, and is equal to P p (Art. 53). For the same reason, the 
resultant of P and p is at C, and equal to P p . But P p = 
P p , and they act in opposite directions. Hence C is at rest, 
and therefore all the forces are in equilibrium. 

64. Equilibrium of Forces in Different Planes. Since 
all the forces which can operate on a body may be reduced to three 
forces on rectangular axes, it is obvious tliat the whole system of 
forces cannot be in equilibrium till the sum of the components on 
each axis is reduced to zero. We must have, therefore, in Art. 59, 
as conditions of equilibrium, these three equations for the three 
axes, X, Y, and Z\ 

P.cosa 4- P .cosa + P".cosn" +, &c., = 0; 
P.cos/34- P .cosjS + P".cos/V +,&c., = 0; 
P. cosy + P .cosy + P".cosy" +,&c., = 0. 

65. Forces Resisted by a Smooth Surface, Whenever 
any forces cause pressure upon a smooth surface, and are held in 
equilibrium by its resistance, the resultant of those forces must be 


at right angles to the surface. Suppose that D A (Fig. 44) 
either a single force or the resultant of 
two or more forces, and that it is held 
in equilibrium by the reaction of A B, 
a smooth surface. If D A is not perpen 
dicular to the surface, it can be resolved 
into two components, one perpendicular 
to the surface A I>, the other parallel to 
it. The former, D B, is neutralized by 
the resistance of the surface ; the latter, B A, is not resisted, and 
produces motion parallel to the surface, contrary to the supposi 
tion. Therefore D A, if held in equilibrium by the surface A B, 
must be perpendicular to it. 



66. The Centre of Gravity Defined. In every body and 
in every system of bodies, there is a point so situated that all the 
parts acted on by the force of gravity balance each other about it 
in every position. That point is called the centre of gravity. The 
force of gravity acts in parallel lines on every particle of a body ; 
the centre of gravity must therefore be the point through which 
the resultant of all these parallel forces is directed, in every posi 
tion of the body. Hence, if the centre of gravity is supported, the 
body is supported. As to the support of the body, therefore, we 
may imagine all parts of it to be collected in its centre of gravity. 
When a system of bodies is considered, they are conceived to be 
united to each other by inflexible rods, which are without weight. 

67. Centre of Gravity of Equal Bodies in a Straight 
Line. The centre of gravity of two equal particles is in the 
middle point between them. Let A and B (Fig. 45), two equal 
particles, be joined by a straight line, and let 

A a and B ~b represent the forces of gravity. 
The resultant of these forces, since they are 
parallel and equal, will pass through the mid 
dle of A B (Art. 54) ; G is therefore the centre 
of gravity. In like manner it is proved that 
the centre of gravity of two equal bodies is 
in the middle point between their respective centres of gravity. 



Any number of equal particles or bodies, arranged at equal 
distances on a straight line, have their common centre of gravity 
in the middle; since the above reasoning applies to each pair, 
taken at equal distances from the extremes. Hence, the centre of 
gravity of a material straight line (e.g., a fine straight wire) is in 
the middle point of its length. 

68. Centre of Gravity of Regular Figures. In the dis 
cussion of the centre of gravity in relation to form, bodies arc con 
sidered uniformly dense, and surfaces are regarded as thin lamina} 
of matter. 

In plane figures the centre of gravity coincides with the centre 
of magnitude, when they have such a degree of regularity that there 
are two diameters, each of which divides the figure into equal and 
symmetrical parts. 

The circle, the parallelogram, the regular polygon, and the 
ellipse, are examples. 

For instance, the regular hexagon (Fig. 46) is divided sym 
metrically by A B, and also by CD. Conceive 
the figure to be composed of material lines 
parallel to A B. Each of these has its centre 
of gravity in its middle point, that is, in C D, 
which bisects them all (Art. 67). Hence, the 
centre of gravity of the whole figure is in C D. 
For the same reason it is in A B. It is, there 
fore, at their intersection, which is also the 
centre of magnitude. 

By a similar course of reasoning it is shown that in solids of 
uniform density, which are so far regular that they can be divided 
symmetrically by three different planes, the centres of gravity and 
magnitude coincide ; e.g., the sphere, the parallelepiped, the cylin 
der, the regular prism, and the regular polyhedron. 

69. Centre of Gravity between Two Unequal Bodies. 

The centre of gravity of two unequal bodies is in a straight line 

joining their respective centres of gravity, and at the point which 

divides their distance in the inverse ratio of their weights. Let 

A a and B h (Fig. 47), passing through the 

centres of gravity of A and B, be proportional 

to their weights, and therefore represent the 

forces of gravity exerted upon them. By the 

laws of parallel forces, the resultant G g = 

A a : + B I (Art. 53), and A a : P> I : : B G : 

A G. Therefore the centre of gravity must be 

at G, through which the resultant passes (Art. 

66). This obviously includes the case of equal weights (Art 67). 


It appears from the foregoing that the whole pressure on a 
support afc is A + B, and that the system is kept in equilibrium 
by such support. 

70. Equal Moments with Respect to the Centre of 
Gravity. If A is put for the weight of A, and B for that of B 9 
the above proportion becomes A : B : : B G : A Gf. 

Let the proportion be changed to an equation, and we have 
A xAG = fixI>G. Suppose now that A B is an inflexible 
rod, without weight, and free to revolve about G. Since the 
bodies balance each other about that point, the equal products, 
A x A G and B x B G, may be taken to represent the equal ten 
dencies of the bodies to turn the system about G. The tendency 
of the body A, expressed by A x A G, is called the moment of J, 
with reference to the point G\ and, similarly, B x B G is the mo 
ment of By with reference to the same point. Hence the proposi 
tion, that the moments of two bodies with reference to their centre 
of gravity are equal. 

71. Centre of Gravity between Three or More 
Bodies. The method of determining the centre of gravity of 
two bodies may be -extended to any number. 

Let A, B } C, D, &c. (Fig. 48), be the weights of the bodies, and 
let the centres of gravity of A and B be con 
nected together by the inflexible line A B. ^ FIG. 48. 

Divide A B so that A : B : : B G : A G, or 
A + B: B::AB : AG , then G is the centre 
of gravity of A and B. Join O G ; and since 
A + B may be considered as at the point G, c D 

divide C G so that A + B + C: C: : C G : G g. In like manner, 
K, the centre of gravity of four bodies, is found by the proportion, 
A + B + C + D: Dii D g\g K. The same plan may be pur 
sued for any number of bodies. 

72. Centre of Gravity of a Triangle. The centre of gravity 
of a triangle is one-third of the distance from the middle of a side 
to the opposite angle. Bisect A C in D (Fig. 49), and B C in E; 
join A E, B D, and D E. B D bisects all lines 

across the triangle parallel to A (7; therefore the 
centre of gravity of all those lines that is, of the 
triangle is in B D. For a like reason, it is in 
A E, and therefore at their intersection, G. Since 
E C=i,BC, SM&D C = A C, . . E D = ^AB. 
But -#(?/? and A G B~B,TQ similar; .-. D G : 
BG::DE:AB::l : 2; ,:D G = B G = 1 BD. 

73. Centre of Gravity of an Irregular Polygon. Divide 
the polygon into triangles by diagonals drawn through one of its 



angles, and then proceed according to the methods already given. 

Let A CE (Fig. 50) be an irregular polygon, whose centre of grav 

ity is to be found. Divide 

it into the triangles P, Q, 

J t \ ,S by diagonals through 

A, and find their centres of 

gravity a, b, c, d (Art, 72). 

Join a b, and divide it so 

tlisitab .a G::P+Q:Q , 

then G is the centre of 

gravity of the quadrilateral 

P + Q. Then join O c, and 

make Gc: Gg::P+Q + R 

: JR. By proceeding in this 

manner till all the triangles 

are used, the centre of gravity of the polygon is found at the last 

point of division. 

74. Centre of Gravity of the Perimeter of an Irregu 
lar Polygon. Find the centre of gravity of each side, which is 
at its middle point, and then 

proceed as in Art. 71, the 

weight of each line being 

considered proportional to 

its length. Thus, let a, b, 

c, &c., be the centres of 

gravity of the sides, A B, 

BC, CD, &c. (Fig. 51); 

join a b, and divide it so 

that a^: a G::AB + B C 

: B C; then G is the centre 

of gravity of A B and B C. 

Next join G c, and make G c 

then g is the centre of gravity of those three sides. 

this manner till all the sides are used. 

The perimeter of a polygon having the degree of regularity de 
scribed in Art. 68, has its centre of gravity at the centre of the 
figure, as may be easily proved. If a polygon has a less degree of 
regularity than that, the centre of gravity both of its area and its 
perimeter may usually be found by methods more direct and 
simple than those given for polygons wholly irregular. 

75. Centre of Gravity of a Pyramid. TJic centre of 
gravity of a triangular pyramid is in the line joining the vertex 
and the centre of gravity of the base, at one-fourth of the distance 
from the base to the vertex. 

A B + B C + CD 

Proceed in 



FIG. 53. 

Let G (Fig. 52) be the centre of gravity of the base BDC: and 
g that of the face ABC. The line A G passes through the cennv 
of gravity of every lamina 
parallel to D B C, on account 
of the similarity and similar 
position of all those laminse ; 
/. the centre of gravity of the 
pyramid is in A G. For a 
similar reason, it is in D g ; 
and therefore at their inter 
section, 0. Now E G = I E D, 
and Eg = J E A ; hence, by 
similar triangles, g G = \ A D. 
But Gg and A OD are also 
similar; .-. G 0=j A 0=\AG. 

From this it is readily proved that the centre of gravity of every 
pyramid and cone is one-fourth of the distance from the centre of 
gravity of the base to the vertex. 

76. Examples on the Centre of Gravity. 

1. A, B, and C (Fig. 53), weigh, respectively, 3, 2, and 1 
pounds, A B = 5 ft., B C = 4 ft., and 

C A = 2 ft. Find the distance of 
their centre of gravity from C. 

First, from the given sides of the 
triangle ABC, calculate the angles. 
A is found to be 49 274 . Next find 
the place of G, the centre of gravity of 
A and B, by the proportion, A + B : B : : A B : A G; A G is 2 
feet, equal to A C. Calculate C G, the base of the isosceles tri 
angle AGO. Its length is 1.673. Then find Cg by the propor 
tion C G : Cg : : A + B + C: A + B\ therefore Cg = 1.394. 

2. A --= 6 Ibs., B = 3 Ibs., and C = 12 Ibs. ; A B = 8 ft, A = 
4 ft., and the angle A is 90 ; find the distance of the centre of 
gravity of A, B, and (7, from C. Ans. 2 ft. 

3. Three equal bodies are placed at the angles of any triangle 
whatever ; show that the common centre of gravity of those bodies 
coincides with the centre of gravity of the triangle. 

4. Find the centre of gravity of five equal heavy particles 
placed at five of the angular points of a regular hexagon. 

Ans. It is one-fifth of the distance from the centre to the 
third particle. 

5. A regular hexagon is bisected by a line joining two opposite 
angles; where is the centre of gravity of one-half? 

Ans. Four-ninths of the distance from the centre to the 
middle of the second side. 


G. A square is divided by its diagonals into four equal part?, <mc.> 
of which is removed ; find the distance from the opposite side of 
the square to the centre of gravity of the remaining figure. 

A us. T 7 g of the side of the square. 

7. Two isosceles triangles are constructed on opposite sides of 
the same base, the altitude of the greater being h, and of the less, 
//; where is the centre of gravity of the whole figure? 

Ans. On the altitude of the greater triangle, at a distance 
from the cprnmon base equal to I (li h ). 

8. The base and the place of the centre of gravity of a triangle 
being given, required to construct the triangle. 

9. Given the base and altitude of a triangle ; required to con 
struct the triangle, when its centre of gravity is perpendicularly 
over one end of the base. 

10. On a cubical block stands a square pyramid, whose base, 
volume, and mass are respectively equal to those of the cube ; 
where is the centre of gravity of the figure ? 

Ans. One-eighth of the height of the cube above its 
upper surface. 

77. Centre of Gravity of Bodies in a Straight Line 
referred to a Point in that Line. If several bodies are in a 
straight line, their common centre of gravity may be referred to a 
point in that line ; and its distance from that point is obtained by 
multiplying each iceight into its own distance from the same point, 
and dividing fie sum of the products by the sum of the weights. 
Let A, By C, and Z>, represent the weights of several bodies, whose 
centres of gravity are in the straight line D (Fig. 54). Eequired 

FIG. 54. 

O A B C D 

the distance of their common centre of gravity from any point 
assumed in the same line. Let G be their common centre of 
gravity, then the moments of A and B must be equal to the op 
posing moments of C and D with reference to the point G (Art. 70). 
That is, 

AxAG + BxBG = CxCG + DxDG:, or, 

A x (QG-OA) + B x (OG-OB) = C x (0 C -OG) + Dx 
(OD- 0); 

expanding, transposing the negative products, and factoring, we 

OB + C x OC+DxOD. 


G = 


A x A + B x OS + Ox C + D x Z> 

78. Centre of Gravity of a System referred to a 
Plane. If the bodies are not in a straight line, they may be re 
ferred to a plane, which is assumed at pleasure. The distance of 
their common centre of gravity from that plane is expressed as 
before: multiply each weight into its oivn distance from the plane, 
and divide the sum of the products by the sum of the bodies. 

Letp,p , p" (Fig. 55), represent the weights of several bodies, 
whose centres of gravity are at those points respectively, and let 
A C be the plane of reference. Join p p , and let g be the com 
mon centre of gravity of p and 
p ; draw p x, g k,p x at right FIG. 55. 

angles to the plane A C, and 
consequently parallel to each 
other ; join x x , and since the 
points p, g, p f , are in a straight 
line, the points x, k, x will also 
be in a straight line, and there 
fore x x will pass through k. 
Join g p", and let G be the com 
mon centre of gravity Qip,p ,p" ; 
draw G K,p" x", perpendicular 
to the plane ; and through g 
draw m n parallel to x x meet 
ing p x produced in n. 

sim. triangles) p m : p n ; 

:.p x p n = p x p m, or p x (n x p x) = p x (p 1 x m x } 


n x = g k = m x f , .-. p x (g k p x)=p r x (p r x g k), 





J > 

, K 

1 \ 

G \ 


for the same reason, if p + p is placed at g > we have 

r K (ff +y ) x 9 b +P" X P" x " __P*P x+p x 

(p + p } + p" p+p +p" 

a formula which is applicable to any number of bodies. 

Let the last equation be multiplied by the denominator of the 
fraction, and we have 

(p+y+^" + &c.) G Kp^p x+p xp x +p"xp" #" + &c.; 
that is, the moment of any system of bodies with reference to a given 


plane, equals the sum of the moments of all the parts of the s/> 
luith reference to the same plane. 

79. Centre of Gravity of a Trapesoid. As an example 
of the foregoing principle, let it be proposed to find the centre of 
gravity of a trapezoid, considered 
as composed of two triangles. The 
centre of gravity of the trapezoid 

A C (Fig. 56) is in E F, which bi- |I 1^ 

sects all the lines of the figure 
parallel to B 0. Suppose G to be 
the centre of gravity of the trape 
zoid ; through G draw K M per 
pendicular to the bases. LetA^J/ _g Kf , 

h,BC=B,AD = b, and join B D. 

The moment of the trapezoid with reference to 

(B + 1) | . G K The moment of the upper triangle is ~ . " I 

<i O 

the moment of the lower triangle is - . -= ; /. (B + b) ^ . G K = 

a o A 

B h h b Ji 2 7 
T- 3 + -2- 3 7i 

_ B + %1> Ti -, B -f 2 h 

tta. 7 Q- P. q 

h 4- b 6 JJ + o 6 

By similar triangles 

of gravity of a trapezoid is on tfie line which bisects the 
parallel bases, and divides it in the ratio of twice the longer plus the 
shorter to twice the shorter plus the longer. 

1. Four bodies, A, B, C, D, weighing, respectively, 2, 3, 6, and 
8 pounds, are placed with their centres of gravity in a right line, 
at the distance of 3, 5, 7, and 9 feet from a given point ; what is 
the distance of their common centre of gravity from that given 
point ; and between which two of the bodies does it lie ? 

Ans. Between C and D ; and its distance from the given 
point 7^ feet. 

2. There are five bodies, weighing, respectively, 1, 14, 2U, 22, 
and 29^- pounds ; a plane is assumed passing through the last 
body, and the distances of the other four from the plane are, re 
spectively, 21, 5, 6, and 10 feet ; how far from the plane is the 
common centre of gravity of the five bodies? Ans. 5 feet. 

[See Appendix for calculations of the place of the centre of 
gravity of curvilinear bodies.] 


80. Centrobaric Mensuration. The properties of the centre 
of gravity furnish a very simple method of measuring surfaces and 
solids of revolution. This method is comprehended in the two 
following propositions, known as the theorems of kildinuo i 

1. If any line revolve about a fixed axis, which is in the plane of 
that line, the SURFACE which it generates is equal to the product 
of the given line into the circumference described ~by its centre of 

Let any line, either straight or curved, revolve about a fixed 
axis which is in the plane of that line; and let /,/ , f",f "> etc., 
denote elementary portions of the line, d, d , d", d ", &c., the dis 
tances of these portions, respectively, from the axis; then the 
surface generated by /, in one revolution, will be 2 TT df; hence 
the surface generated by the whole line will be 

8 = 2 TT (df + d f + d"f" + d" f" + &c.) . . . (1). 

Put L = the length of the revolving line, and G = the dis 
tance from the axis to the centre of gravity of the line ; then 
(Art. 78) 

GL = df + d f + d"f" + d" f" + &c (2). 

Combining (1) and (2), we have 

#=2trj0Z r ..(i). 

2. If a plane surface, of any form whatever, revolve about a fixed 
axis which is in its own plane, the VOLUME generated is equal to 
the product of that surface into the circumference described by its 
centre of gravity. 

Let any plane surface revolve about an axis which is in the 
plane of that surface; and let /,/ , /",/ ", &c., denote elementary 
portions of the surface, d, d 1 , d", d ", &c., the distances of these 
portions, respectively, from the axis ; then the volume generated 
by /in one revolution will be 2 TT df] hence the volume generated 
by the whole surface will be 

F= 2 TT (df 4- d f + d"f" 4- d "f" + &c.) . . . (4). 

Fut A the area of the revolving surface, and G = the dis 
tance from the axis to the centre of gravity of that surface; then 
(Art 78) 

AG=df + d f + d"f" + d "f" + &c., . . . (5). 
Substituting in (4), we have 

V=%7rA G...(G). 

As an illustration of the first theorem, the straight line C D 
(Fig. 57), revolving about the centre C, describes a circle whose 
surface is equal to (7 7) into the circumference of the circle de 
scribed by its centre of gravity, E. This is evident also from the 


consideration that, since E is the centre of the line C D, the cir 
cumference described by it will -be half the 
length of the circumference A D B ; and the FIG. 57. 

area of a circle is equal to the product of the 
radius into half the circumference. 

The second theorem is illustrated by the 
volume of a cylinder, whose height = h, and 
the radius of whose base = r. 

Common method ; base = TT r 3 ; height li ; 
/. yol. = TT r 2 h. 

Centrobaric method ; revolving area = r h ; circumference de 
scribed by the centre of gravity = r x 2 rr ; .. vol. = r li . \ r . 
27T = rrr*h. 

81. Examples. 

1. Suppose the small circle (Fig. 57) to be placed with its 
plane perpendicular to the plane of the paper, and revolved about 
C, the point D describing the line D B A ; required the content 
of the solid ring. If C D = R, and E D = r, then the area re 
volved = TT r 3 , and the circumference D B A = 2 TT R ; /. the ring 
2 TT 2 R r 3 . It is equal to a cylinder whose base is the circle 
E D, and whose height equals the line DBA. 

2. Find the convex surface of a cone ; slant height = s ; and 
rad. of base = r. The line revolved being s, and the distance 
from the axis to its centre of gravity, ^ r, the surface is TT r s. 

3. A square, whose side is one foot, is revolved about an axis 
which passes through one of its angles, and is parallel to a diago 
nal ; required the volume of the figure thus formed. 

Am. TT V2, or 4.4429 cubic ft. 

4. Find the surface of a sphere whose radius is r. (The dis 
tance from the centre of a circle to the centre of gravity of its 

semi-circumference is . Appendix, Art. 92.) 

2 r 

Ans. TT r . . 2 TT = 4 TT r 1 . 


5. Find the volume of a sphere whose radius is r. (Appendix, 
Art. 95.) % Ans. | TT r 3 . 

82. Support of a Body. A body cannot rest on a smooth 
plane, unless it is horizontal; for the pressure on a plane (Art. 0." ) 
cannot be balanced by the resistance of that plane, except when 
perpendicular to it; therefore, as the force of gravity is vertical, 
tlv resisting plane must be horizontal. 

The base of support is that area on the horizontal plane which 
is comprehended by lines joining the extreme points of contact. 


FIG. 58. 

If there are three points of contact, the base is a triangle ; if 
four, a quadrilateral, &c. 

When the vertical through the centre of gravity (called the 
line of direction) falls with 
in the base, the body is 
supported ; if without, it is 
not supported. In the body 
A (Fig. 58) the force of 
gravity acts in the line G F, 

and there are lines of re- ^ 

si stance on both sides of 

G F, as G C and G E, so that the body cannot turn on the edge 
of the base, without rising in an arc whose radius is G C or G E. 
But, in the body B, there is resistance only on one side ; and 
therefore, if the force of gravity be resolved on G C and a perpen 
dicular to it, the body is not prevented from moving in the direc 
tion of the latter, that is, in the arc whose radius is G C. 

If the line of direction fall at the edge of the- base, the least 
force will overturn it. 

83. Different Kinds of Equilibrium. If the base is re 
duced to a line or point, then, though there may be support, there 
is no firmness of support ; the body will be moved by the least 
force. But it is affected very differently in different cases. 

When it is moved from its position of support and left, it will 
in some cases return to it, pass by, and return again, and continue 
thus to vibrate till it settles in its place of support by friction and 
other resistances. This condition is called stable equilibrium. 

In other cases, when moved from its position of support and 
left, it will depart further from it, and never recover that position 
again. This is called unstable equilibrium. 

In other cases still, the body, when moved from its place of 
support and left, will remain, neither returning 
to it nor departing further from it. This is called FlG - 

neutral equilibrium. JB 

84. Stable Equilibrium. Let the body 
(Fig. 59) be suspended on the pivot A. This is 
its base of support. While the centre of gravity 
is below A, the line of direction EOF passes 
through the base, and the body isr supported. Let 
it be moved aside, and the centre of gravity be 
left at G. Let G R represent the force of gravity, 
and resolve it into G N- on the line A G, and NR, 
or G B, perpendicular to A G. G N is resisted 
by the strength of A, and G B moves the centre 



FIG. 60. 

of gravity in the arc whoso radius is A G. Hence the body swings 
with accelerated motion till the centre of gravity reaches 0, where 
the force G B becomes zero. But by its inertia, the body passes 
beyond that position, and ascends on the other side, till the retard 
ing force of gravity stops it at g, as far from as G is. It then 
descends again, and would never cease to oscillate were there no 

85. Unstable Equilibrium. Next, let the body be turned 
on the pivot till the centre of gravity G is at P, above A (Fig. GO). 
Then, as well as when G is below A, the body is 

supported, because the line of direction E P F passes 
through the base A. But if turned and left ;n the 
slightest degree out of that position, it cannot re 
cover it again, but will depart further and further 
from it. Let R represent the force of gravity, 
and let it be resolved into G N, acting through A 9 
and G B perpendicular to it. The former is re 
sisted by A ; the latter moves G away from P, the 
place of support. If the body is free to revolve 
about A, without falling from it, the centre of 
gravity will, by friction and other resistances, finally 
settle below A, as in the case of stable equilibrium. 

86. Neutral Equilibrium. Once more, suppose the pivot 
supporting the body to be at G, the centre of gravity ; then, in 
whatever situation the body is left, the line of direction passes 
through the base, and the body rests indifferently in any position. 

These three kinds of equilibrium may be illustrated also by 
bodies resting by curved surfaces on a horizontal plane. Thus, if 
a cylinder is uniformly dense, it will always have a neutral equi 
librium, remaining wherever it is placed. But if, on account of 
unequal density, its centre of gravity is not in the axis, then its 
equilibrium is stable, when the centre of gravity 
is below the axis, and unstable when above it. 

In general, there is stable equilibrium when 
the centre of gravity, on being disturbed in 
either direction, begins to rise ; unstable when, 
if disturbed either way, it begins to descend; 
and neutral when the disturbance neither raises 
nor lowers the centre of gravity. 

87. Questions on the Centre of Grav- 

1. A frame 20 feet high, and 4 feet in diam 
eter, is racked into an oblique form (Fig. 01), B F c 


till it is on the point of falling ; what is its inclination to the 
horizon? Ans. 78 27 47". 

2. A stone tower, of the same dimensions as the former, is in 
clined till it is about to fall, but preserves its rectangular form ; 
what is its inclination ? Ans. 78 41 24". 

3. A cube of uniform density lies on an inclined plane,- and is 
prevented by friction from sliding down ; to what inclination 
must the plane be tipped, that the cube may just begin to roll 
down ? Ans. 45. 

4. What must be the inclination of a plane, in order that a 
regular prism of any given number of sides may be at the limit 
between sliding and rolling down ? 

Ans. Equal to half the angle at the centre of the prism ? 
subtended by one side. 

5. A body weighing 83 Ibs. is suspended, and drawn aside from 
the vertical 9 ; what pressure is there on the point of support, 
and what force urges it down the arc ? 

Ans. Pressure on the support, 81.978 Ibs. 
Moving force, 12.984 Ibs. 

88. Motion of the Centre cf Gravity of a System 
when one of the Bodies is Moved. 

When one body of a system is moved, the centre of gravity of the 
system moves in a similar path, and its velocity is to that of the moving 
body as the mass of that body is to the mass of the whole system. 

If the system contains but two bodies, A and B (Fig. 62), sup 
pose A to remain at rest, while B 
describes the straight lines B C, 
C D, &c., the centre of gravity G 
will in the same time describe the 
similar series, G H, H J, &c. 
When B is in the position B, and 
the centre of gravity at G, A G : 
A B : : B : A -f B\ when B is 
at (7, A H : A C : : B : A + J3-, 
:.AG\AB-. .AHiAC. Hence 
G H is parallel to B C, and G H\ B C: : B : A + B. In like 
manner, HJ : CD : : E : A + B, &c. Thus, all the parts of one 
path are parallel to the corresponding parts of the other, and have 
a constant ratio to them. Therefore the paths are similar. As 
the corresponding parts are described in equal times, their lengths 
are as the velocities. But the lengths are as B : A + B\ there 
fore the velocity of the common centre of gravity is to that of the 
moving body as the mass of the moving body is to the mass of both. 
The same reasoning is applicable when the body moves in a curve. 


If the system contain any number of bodies, and the centre of 
gravity of the whole be at G, then the centre of gravity of all 
except B must be in the line B G beyond G. Suppose it to be at 
A, and to remain at rest, while B moves; then it is proved in the 
same manner as before, that G, the centre of gravity of the whole 
system, moves in a path parallel to the path of B, and with a 
velocity which is to B s velocity as the mass of B to the mass of 
the entire system. 

89. Motion of the Centre of Gravity of a System when 
Several of the Bodies are Moved. 

When any or all of the bodies of a system are moved, the centre 
of gravity moves in the same manner as if all the system lucre collected 
there, and acted on by the forces which act on the separate bodies. 

Let A, By (7, &c. (Fig. G3), belong to a system containing any 
number of bodies, and let M be the mass of the system. Let A 
be moved over A a, B over B b, C over C c, &c. And first sup 
pose the motions to be made in equal successive times. If the 
centre of gravity of the system is first at G, then that of all the 
bodies except A is in A G produced, as at g. While A moves to , 
G moves in a parallel line to H (Art. 88), and G H : A a : : A : M. 
In like manner, when B describes B b, the centre of gravity of the 
other bodies being at h, the centre of gravity of the system de 
scribes the parallel line, // /i, and H K : B b : : B : M\ and when 
C moves, K L : C c :: C : J/, &c. Now, A a and G H represent 
the respective velocities of the 
body A, and the system M\ FIG. 63. 

therefore, if we convert the 
proportion G H : A a : : A : M 
into an equation, we have A x 
A a = M x G H] that is, the 
momentum of the body A 
equals the momentum of the 
system M. It therefore re 
quires the same force to move 
A over A a as to move the system M over G H. The same is true 
of the other bodies. If then the several forces which move the 
bodies, limiting the number to three, for the present, were applied 
successively to the system collected at G, they would move it over 
G H, H K, K L. But if applied at once, they would move it over 
G Z, the remaining side of the polygon. If, therefore, the forces, 
instead of acting successively on the bodies, were to move A over 
A. a, B over B b, and G over C c, at the same time, the centre of 
gravity of the system would describe G Lin the same time. In 11 > /.- 
same way it may be proved, that whatever forces are applied to the 


several bodies of a system, the centre of gravity of the system is 
moved in the same manner as a body equal to the whole system 
would be moved, if all the same forces were applied to it. 

It is possible that the centre of gravity of a system should 
remain at rest, while all the bodies in it are in motion. For, sup 
pose all the forces acting on the bodies to be such that they might 
be represented in direction and intensity by all the sides of a poly 
gon, then, since a single body acted on by them would be in equi 
librium, therefore the centre of gravity of the system would remain 
at rest, though the bodies composing it are in motion. 

90. Mutual Action among the Bodies of a System. 

The forces which have b een supposed to act on the several bodies 
of a system are from without, and not forces which some of the 
bodies within the system exert on others. If the bodies of a sys 
tem mutually attract or repel each other, such action cannot affect 
the centre of gravity of the whole system. For action and reac 
tion are always opposite and equal. Whatever force one body 
exerts on any other to move it, that other exerts an equal force on 
the first, and the two actions produce equal and opposite effects on 
the centre of gravity between them. Therefore the centre of 
gravity of a system remains at rest, if the bodies which compose it 
are acted on only by their mutual attractions or repulsions. 

91. Examples on the Motion of the Centre of Gravity. 

1. Two bodies, A and B, of given weights, start together from 
D (Fig. 64), and move uniformly with given velocities in the direc 
tions D A and D B ; required the di 
rection and velocity of their centre of FIG. 64. 

As the directions of D A and D B 
are given, we know the angle A D B ; 
from the given velocities, we also know 
the lines D A and D B, described in a 
certain time. Calculate the side A B, 
and the angles A and B. Find the 

place of the centre of gravity G between the bodies at A and B. 
Then, in the triangle D B G, D B, B G, and angle B are known, 
by which may be found the distance D G passed over by the cen 
tre of gravity in the time, and B D G the angle which its path 
makes with that of the body B. 

2. The bodies A and B, of given weights, start together from D 
(Fig. 65), and move with equal velocities in opposite directions 
around the circumference of a circle, meeting again at D ; what is 
the path of their centre of gravity ? 

Draw the diameter, D E, and join A and B, the points which 



the bodies have reached after any given time. As D A and I) l> 
are equal arcs, A B is perpendicular to D E> and is bisected by it 
Let G be the common centre of gravity of A and B, then 

A:B:- B G-.A G; 

.-. A+ B .A-B-.BG + AG-.B G-AG; 
:: AB : Q M\ 


FIG. 65. 

Therefore A N, the ordinate of the circle, is to G N, the corre 
sponding ordinate of the figure de 
scribed by the centre, always in the 
same constant ratio, of the sum of 
the bodies to their difference. But 
this is a property of the ellipse, that, 
when its axis is the diameter of a 
circle, the corresponding ordinates 
of the two figures are in a constant 
ratio. Hence the centre of gravity 
of A and B describes an ellipse, 
while they move, in the manner 
before stated, round the circle. 

If the bodies approach equality, 

their difference grows less, and therefore the ellipse more eccentric, 
till, when the bodies are equal, the path of the centre of gravity is 
a straight line, as it evidently should be, in order to bisect the 
chords, A B, 1) a, &c. 

3. Three bodies of given weight, A,B, C, in the same time and 
in the same order, describe with uniform velocity the three sides 
of the given triangle A B C (Fig. 66) ; required the path of their 
centre of gravity. 

Let G be their centre of 
gravity before they move. If 
they move successively, G de 
scribes G /i, K L, L M, par 
allel to the sides of the trian- 

FIG. 66. 

gle, and having to them re 
spectively the same ratios as 

the corresponding moving 

bodies have to the sum of the C L K 

bodies (Art, 89). Thus, three 

sides of the polygon are known ; and the angle KB^ and L = C. 
These data are sufficient for calculating the fourth side, G M, which 
the centre of gravity describes, when the bodies move together. 

4. Show that when the three bodies in Example 3d are equal, 
the centre of gravity will remain at rest. 



5. A (Fig. 67) weighs one pound; 
B weighs two pounds, and lies direct 
ly east of Ay they move simulta 
neously, A northward, and B east 
ward, at the same uniform rate of 40 
feet per second ; required the direc 
tion and velocity of their centre of 

FIG. 67. 

Ans. Velocity is 29.814 feet per second. 
Direction is E. 26 33 54" 1ST. 



92. Elastic and Inelastic Bodies. Mastic bodies are 
those which, when compressed, or in any way altered in form, 
tend to return to their original state. Those which show no such 
tendency are called inelastic or non-elastic. No substance is 
known which is entirely destitute of the property of elasticity ; 
but some have it in so small a degree, that they are called inelas 
tic, such as lead and clay. Elasticity is perfect when the restoring 
force, whether great or small, is equal to the compressing force. 
Air, and the gases generally, seem to be almost perfectly elastic ; 
ivory, glass, and tempered steel, are imperfectly, though highly, 
elastic ; and in different substances, the property exists in all con 
ceivable degrees between the above-named limits. 

93. Mode of Experimenting. Experiments on collision 
may be made with balls of the 

same density suspended by FIG. 68 

long threads, so as to move in 
the line which joins their cen 
tres of gravity. If the arcs 
through which they swing are 
short compared with their 
radii, the balls, let fall from 
different heights, will reach the 
bottom sensibly at the same 
time, and will impinge with 
velocities which are very nearly 
proportional to the arcs. Thus 
A (Fig. 68), falling from 6, 


and B from 3, will come into collision at 0, with velocities which 
are as 2 : 1. 

94. Collision of Inelastic Bodies. Such bodies, after im 
pact, move together as one mass. 

The velocity of two inelastic bodies after collision is equal to the 
algebraic sum of their momenta, divided by the sum of the bodies. 

Let A 9 B, represent the masses of the two bodies, and a, b, their 
respective velocities. Considering a as positive, if B moves in the 
opposite direction, its velocity must be called b. Let v be the 
common velocity after impact. 

1. Same directions. The momentum of A is A a ; that of B is 
B b] and the momentum of both after collision is (A + B) v. 
According to the third law of motion (Art. 13), whatever mo 
mentum A loses, B gains, so that the whole momentum is the 
same after collision as before ; therefore 

D-L f A A a -\- B b 

Aa + Bb = (A + B)v,. .v = - . 

JL -f- Jj. 

If B is at rest before impact, b= 0, and v = 

To find the loss or gain of velocity for either body, multiply 
the other body by the difference of velocities, and divide fey the sum 
of the bodies. For, A s velocity before impact was a ; after impact 
. A a + B b A a + B b B (a - b) 

lt 1S - theref re the l SS = a ~ -- 

But B s gain is the velocity after impact diminished by the velo- 

, - . Aa + Bb A(a b) 

city before, Le.,^-^ - b = -A--^. 

When B is at rest, these expressions become 

Ba f ., , A a , 

~2 - j> f r -4 lss ; -j - for B s gam. 

2. Opposite directions. Since b is negative, v = A - ^ . 

-A. -f- B . 

To find loss QI gain in this case, multiply the other body by the 
sum of the velocities, and divide by the sum of the bodies. For, 
A s loss 

-^_b _ B(a + b) 


A+B A + B > 
and B s gain 

Aa-Bb T\_Aa-Bb , A 

: - 

A + B A+B 

In the case of opposite motions, the formula for v becomes zero, 
when A a B b ; but in that case, A : B : : b : a. Hence, if 
bodies which meet each other have velocities inversely as the 
quantities, they will be at rest after the collision. 


95. Questions on Inelastic Bodies. 

.1. A, weighing 3 oz., and moving 10 feet per second, overtakes 
B, weighing 2 oz., and moving 3 feet per second ; what is the 
common velocity after impact? Ans. 7 5 feet per second. 

2. A weight of 7 oz., moving 11 feet per second, strikes upon 
another at rest weighing 15 oz. ; required the velocity after im 
pact ? Ans. 3 J feet per second. 

3. A weighs 4 and B 2 pounds ; they meet in opposite direc 
tions, A with a velocity of 9, and B with one of 5 feet per second ; 
what is the common velocity after impact ? 

Ans. 4J feet per second. 

4. A ^ pounds, B 4 pounds ; they move in the same di 
rection, with velocities of 9 and 2 feet per second; required the 
velocity lost by A and gained by B? Ans. A 2 T 6 T , B 4 T 5 T . 

5. A body moving 7 feet per second, meets another moving 
3 feet per second, and thus loses half its momentum; what 
are the relative masses of the two bodies ? 

Ans. A:B::13:7. 

6. A weighs 6 pounds and B 5 ; B is moving 7 feet per sec 
ond, in the satae direction as A ; by collision B s velocity is 
doubled; what was A s velocity before impact? 

Ans. 19 1 feet per second. 

96. Collision of Elastic Bodies. Elastic bodies after col 
lision do not move together, but each has its own velocity. These 
velocities are found by doubling the loss and gain of inelastic 
bodies. When the elastic body A impinges on B, it loses velocity 
while it is becoming compressed^ and again, while recovering its 
form, it loses as much more, because the restoring force is equal to 
the compressing force. For a like reason, B gains as much velo 
city while recovering its form as it gained while being compressed 
by the action of A. Hence, doubling the expressions for loss and 
gain given in Art. 94, and applying them to the .original velocities, 
we find the velocity of each body after collision, on the supposition 
of perfect elasticity. 

When the directions are the same, 

2 B (a - I) 
the velocity of A ~ a j- v ^~- ; 

A. + > 

ru P T> T. 2 A (a I) 

that of B = I + T \ \ 
-4 + tf 

When the directions are opposite, 

,, 2B(a + fy 
the velocity of A = a ~ ^~- ; 

,, 2 A (a + 5) 

that of B I + - v -. 


Reducing these expressions, we have for the velocities of elastic 
bodies after collision the following formulae : 

(1.) Same direction. Velocity of A = ^1^M^- 

^1 ~\~ JO 

,. TT -I -i. f -D 
(2.) Same direction, Velocity of B = 


(3.) Opposite directions. Velocity of A = - -^- j~ . 

jA. + JB 

(4.) Opposite directions. Velocity of B = - ^ . 

97. Equal Elastic Bodies. After the impact of equal elastic 
bodies, each takes the original velocity of the other. "When A = B, 
formula (1) is reduced to b; and formula (2) to #; that is, A has 
B s former velocity, and B has A s. The same is true if they move 
in opposite directions. For, when A = B, formula (3) becomes 
b, which was B s original velocity, and formula (4) becomes a, 
which was A s. Therefore, in the opposite motions of equal elastic 
bodies, collision causes each to rebound, since + a is exchanged 
for b, and b for + a. 

If we reduce these four formulae for the case in which A = B, 
and B is at rest, we find the same interchange of conditions ; for 
formula (1) becomes 0, and formula (2) becomes a\ so formula (3) 
becomes 0, and formula (4) becomes a. 

98. Unequal Elastic Bodies. 

1. If a greater body impinge on a less one at rest, the imping 
ing body goes forward, but slower than before, and the other pre 
cedes it with a greater velocity than the impinging body first had. 

For, formula (1) becomes ^3 -- -~ 9 which is positive, but less 

jA. -f~ Jo 

than a; therefore it advances, though slower than before. But 

2 A a 

formula (2) becomes -7-^^, which is greater than a; hence, B 

goes on faster than A did before collision. 

2. If a less body impinge on a greater one at rest, it rebounds, 
and the other goes forward, but with a less velocity than that 

which the impinging body first had. For, J. is negative, 

^1 ~{~ JO 

, ZAa . , 
and - - ^ is less than a. 

A. ~\~ JO 

3. If two elastic bodies, having equal velocities, meet each 
other, and one of them is brought to rest, its mass is three times 
as great as that of the other. For. as the velocities are equal, by 


formula (3), 

= 0; ... (A - B) a - 2 Ba = 0; 

FIG. 69. 

99. Series of Elastic Bodies. 

1. Equal bodies. Let a row of equal elastic bodies, A, B, C. . . 
(Fig. 69) be suspended in contact; then 

(Art. 97), if A be drawn back and left to 
fall against B, it will rest after impact, and 
E will tend to move on with A s velocity ; 
after the impact of B on C, B will remain, 
and C 9 tend to move with the same velocity ; 
and so the motion will be transmitted through 
the series, and F will move away, while all 
the others remain at rest. 

2. Decreasing series. If the bodies de 
crease, as A, B, (7, &c. (Fig. 70), and A be 
drawn back to A , and allowed to fall against 
B, then (Art. 98) A still moves fonvard, 
while B receives a greater velocity than A 

had, C still greater, &c. The last of the series, therefore, moves 
with the greatest velocity, and each one with a greater velocity 
than that which impinged on it. 

FIG. 70. 

3. Increasing series. If the bodies increase, as A, B, C, &c. 
(Fig. 71), then, when A falls from A against B, it imparts to B a 

FIG. 71. 

less velocity than it had itself, and rebounds (Art 98) ; in like 
manner B rebounds from (7, and so on ; while the last of the series 
goes forward with less velocity than any previous one would hare 
had if it had been the last. 


If the bodies in Fig. 70 are in geometrical progression, the 


Let the series be A, Ar, Ar* .... Ar n ~ \ 

By Art. 98, when A impinges on B at rest the velocity com- 

2 A a 2 A a 2 a 

mumcated to B is -; - = -- = b. 

A + B A-\- Ar 1 + r 

Again, the velocity imparted to C is 

2Bb 2 Ar 2 a 2* a 

B + C ~ Ar + Ar* X iTr ~~ (T+7J 5 

, ... 2 a Wo, 

Hence the successive velocities are #, - , -^ ^, &c., from 

1 + r (1 + r) 

which it appears that any term in the series is found by multiply 
ing the original velocity by 2, raised to a power one less than the 
number of terms and divided by 1 4- r raised to the same power. 

2 n1 (i 
Consequently, the last term is . ; T . Hence, vel. of the first : 

2 n ~ l (i / 2 \ n - J 

vel. of the last : : a : j- - -. : : I : ( ) 

(1 + r) n ~ \1 + rl 

100. Questions on Elastic Bodies. 

1. A, weighing 10 Ibs. and moving 8 feet per second, impinges 
on B, weighing 6 Ibs. and moving in the same direction, 5 feet 
per second ; what are the velocities of A and B after impact ? 

Ans. A s = 5|, B s = 8|. 

2. A : B : : 4 : 3 ; directions the same ; velocities 5:4; what is 
the ratio of their velocities after impact ? Ans. 29 : 36. 

3. A, weighing 4 Ibs., velocity 6, meets B, weighing 8 Ibs., 
velocity 4 ; required their respective directions and velocities after 
collision ? Ans. A is reflected back with a velocity of 7], 

and B with a velocity of 2~. 

4. A and B move in opposite directions ; A equals 4 B, and 
b = 2 a ; how do the bodies move after collision ? 

Ans. A returns with J, B with 1 its original velocity. 

5. There are ten bodies whose masses increase geometrically 
by the constant ratio 3, and the first impinges on the second 
with the velocity of 5 feet per second ; required the motion of the 
last body ? Ans. The last body would move with the 

velocity of 5 f 2 feet per second. 

101. Living Force lost in the Collision of Inelastic 
Bodies. The amount of living force (Art. 35) before collision is 

A a? 4- B b* ; and after collision it is (A + B) x - 

\ ~^~ / 

. Subtract the latter from the former, and call the 


remainder d. Then d A tf + B V - - . Expanding 

./i ~p x> 

and uniting terms, d = - -^ n~* ^-^^ s va ^ ue f ^ * s positive, 

U3. -p -tt 

because (a 1>Y is necessarily positive, as well as A and B. There 
fore there is always a loss of living force in the collision of inelas 
tic bodies. 

102. Living Force Preserved in the Collision of Elastic 
Bodies. The living force of A before collision is A 2 ; after col 

lision, it is A x j~ p-r 5 . Subtracting the latter from 
(A + ti) 

the former, the loss (supposing there is loss) is 

(A + BY Aa i -(A-BYAa i -^(A B) A B al>-AB* P 

(A + BY 
The living force of B before collision is B 1? ; after collision, it 

{(B A] I 4- 2 Act}* 
is B x - ~~W\* --- > a expression for loss is 

(A + BYB^-(B-AYBy--4,(B-A)ABab-4 : A^Ba 2 

(A + B? 

Therefore the total loss of living force is the sum of the expres 
sions (1.) and (2.). 

Reducing the two first terms in each fraction to one, the frac 
tions become 

(A + BY 

and (A + B) . - .... (4.) 

If the fractions (3) and (4) be add?d, it is evident that the nu 
merators cancel each other, and therefore the sum of the fractions 
is zero. Hence, there is no loss of living force in the collision of 
elastic bodies. 

103. Impact on an Immovable Plane. If an inelastic 
body strikes a plane perpendicularly, its motion is simply destroyed; 
in strictness, however, it imparts an infinitely small velocity to the 
body called immovable. If it strikes obliquely, and the plane is 
smooth, it slides along the plane with a diminished velocity. Let 
A L (Fig. 72) represent the motion of the body before impact on 
the plane P N, and resolve it into 
A C, perpendicular, and C L, par- FlG - 

allel to the plane. Then A C, as 
before, is destroyed, but C L is 
not affected; hence the former 
velocity is to its velocity on the 
plane, as A L : C L : : radius : co 
sine of the inclination. 


If a perfectly clastic body impinges perpendicularly upon a 
plane, then, after its motion is destroyed, the force by which it 
resumes its form causes an equal motion in the opposite direction ; 
that is, the body rebounds in its own path as swiftly as it struck. 
But if tfie impact is oblique, the body rebounds at an equal angle 
on the opposite side of the perpendicular. For, resolve A L, as 
before, into A C,CL; the latter continues uniformly ; but, instead 
of the component A C, there is an equal motion in the opposite 
direction. Therefore, if L D is made equal to C L, and D E equal 
to A C, the resultant of L D and D E is L E, which is equal to 
A L, and has the same inclination to the plane. Hence, the 
angles of incidence and reflection are equal, and on opposite sides 
of the perpendicular to the surface at the place of impact. 

104. Imperfect Elasticity. The formulae for the velocity 
of bodies after collision, and the statements of the preceding arti 
cle, are correct only on the supposition that bodies are, on the one 
hand, entirely destitute of elasticity, or on the other perfectly 
elastic. As no solid bodies are known, which are strictly of either 
class, these deductions are found to be only near approximations 
to the results of experiment. In all practical cases of the impact 
of movable bodies, the loss and gain of velocity are greater than if 
ihey were inelastic, and less than if perfectly elastic. And in cases 
of impact on a plane, there is always some velocity of rebound, but 
less than the previous velocity ; and therefore, if the collision is 
oblique, the body has less velocity, and makes a smaller angle with 
the plane than before. For, making D F less than A C, the 
resultant L F is less than A L, and the angle D L F is smaller 
than D L E, or A L C. 



105. Classification of Machines. In the preceding chap 
ters, the motion of bodies has been supposed to arise from the im 
mediate action of one or more forces. But a force may produce 
effects indirectly, by means of something which is interposed for 
the purpose of changing the mode of action. These intervening 
bodies are called, in general, machines ; though the names, tools, 
instruments, engines, &c., are used to designate particular classes 
of them. The elements of machinery are called simple machines. 
The following list embraces those in most common use : 

1. The lever. 

2. The wheel and axle. 



3. The pulley. 

4. The rope machine. 

5. The inclined plane. 

6. The wedge. 

7. The screw. 

8. The knee-joint. 

In respect to principle, these eight, and all others, may be re 
duced to three. 

1. The law of equal moments, applicable in those cases in which 
the machine turns on a pivot or axis, as in the lever and the wheel 
and axle. 

2. The principle of transmitted tension, to be applied wherever 
the force is exerted through a flexible cord, as in the pulley or 
rope machine. 

3. The principle of oblique action, applicable to all the other 
machines, the force being employed to balance or overcome one 
component only of the resistance. 

The force which ordinarily puts a machine in motion is called 
the power; the force which resists the power, and is balanced or 
overcome by it, is called the weight. 

A compound machine is one in which two or more simple ma 
chines are so connected that the weight of the first constitutes the 
power of the second, the weight of the second the power of the 
third, &c. 


106. The Three Orders of Straight Lever. The lever is 
a bar of any form, free to turn on a 
fixed point, which is called the fulcrum. FIG. 73. ^ ^ 

In the first order of lever, the fulcrum ~ ^ . 

is between the power and weight (Fig. P 11 F 

73) ; in the second, the weight is between @ 

the power and fulcrum (Fig. 74) ; in 

the third, the power is between the weight and fulcrum (Fig. 75). 

FIG. 74 FIG. 75. 

If P and IF, in either of these figures, represent forces acting 


in vertical lines, then the circumstances of equilibrium are deter 
mined by the laws of parallel forces (Art. 54). In Fig. 73, if P 
and W are in equilibrium, their resultant will be so situated at (\ 
that P : W : : B C : A C\ and the fulcrum must be at that point, 
and be able to sustain a pressure equal to P -f- W. In Fig. 74, P 
and the reaction of F at C are two upward forces, whose resultant 
is counterbalanced by IF; then W is represented by the whole 
line, and P by the part B G\ . . P : W : : B C : A C, as before. 
The pressure on F equals W P. In Fig. 75, W and the reac 
tion of F are downward forces, whose resultant is at A, in equi 
librium with P. Here P is represented by the whole line B C, 
and W by the part A (7; .-. P : W : : B C : A C. The upward 
pressure against F is equal to P TF. 

Hence, in each order of the straight lever, when the forces act 
in parallel lines, 

TJie power and weight are inversely as the lengths of the arms 
on which they act. 

107. Equal Moments in Relation to t"ie Fulcrum. 
Changing the proportion into an equation, we find for each order 
of the lever, P x A C = W x B C\ that is, 

The power and iveight have equal moments in relation to the 

The moment of either force is the measure of its efficiency to 
turn the lever ; for, since the lever is in equilibrium, the efficiency 
of the power to turn it in one direction must equal the efficiency 
of the weight to turn it in the opposite direction. We may there 
fore use P x A C to represent the former, and W x B C, the 

If several forces, as in Fig. 76, are in equilibrium, some tending 

FIG. 70. 
A B c n 

to turn the bar in one direction, and others in the opposite, then 
A and B must have the same efficiency to produce one motion as 
C and D have to produce the opposite ; that is,AxAG + xBG 
= Ox CG + D x DG\ or, 

The sum of the moments of A and B equals the sum of the mo 
ments of C and D. 

In order to allow for the influence of the weight of the lever 
itself, consider it to be collected at its centre of gravity, and add 
its moment to that of the power or weight, according as it aids 
the one or the other. In Fig. 73, let the weight of the lever = w 9 


and the distance of its centre from C on the side of P = m; then 
P x A O + m w = W x (7. In the 2d and 3d orders, the mo 
ment of the lever necessarily aids the weight ; and hence, in each 
case, P x A C = W x B G + m w. 

If a weight hangs on a bar between two supports, as in Fig. 77, 
it may be regarded as a lever of the 
2d order, the reaction of either sup- Fm - 77> 

port being considered as a power, j 
Let F denote the reaction at A ., and F \ 

F at (7; then by the theorems of 
parallel forces, we have the pressures 

at A and C inversely as their dis- ||| ||TV 

tances from B, and W = F + F . 

108. The Acting Distance. In the three orders, as above 
described, the equilibrium is not destroyed by inclining the lever 
to any angle whatever with the horizon, provided the centre of 
motion C is at the centre of gravity of the bar, and not above or 
below it, and provided the directions of the forces remain vertical. 
For, by the principle of parallel forces, any straight line intersect 
ing the lines of the forces is divided by the line of the resultant 
into -parts which are inversely as the forces; therefore (Fig. 78) 
I C : a C : : P : W. Hence, the re 
sultant of P and W remains at C, in 

every position of the lever. By sim- 

ilar triangles, I C: a C: : CN: C M; 

. .PiWi: G N: CM.-, .-. P x CM 

= W x C N. The lines C M and 

C N, which are drawn from the ful- 

crum perpendicular to the lines in 9 

which the forces act, are called the 

acting distances of the power and 

weight, respectively. And as they may be employed in levers of 

irregular form, the moments of power and weight are usually 

measured by the products, P x CM and W x C N; therefore, the 

power multiplied by its acting distance equals the iveight multiplied 

by its acting distance ; or, more briefly, the moment of the power 

equals the moment of the weight, as in Art. 107. In Figs. 73, 74, 

and 75, the acting distances are in each case identical with the 

arms of the lever. 

109. Lever not Straight, and Forces net Parallel. 
Let ACS (Fig. 79) be a lever of any form, and let it be in equi-,, 
librium by the forces P and P , acting in any oblique directions 
in the same plane. Produce P A and P B till they meet in D ; 
then, if the fulcrum is at C, the resultant must be in the direction 



FIG. 79. 

/) C\ otherwise the reaction of the fulcrum cannot keep the sys 
tem in equilibrium (Art. 60, 2). 
Therefore (Art. 61) P : P : : sin 
B D C : sin A D C. 

Draw C M perpendicular to 
A D, and C N to B D, and they 
are the sines of A D C and B D C, 
to the same radius D C. 
.-.P:P :: CN: CM ,tm&PxCM 

= P x CN. 

The lines C M and C N are the act 
ing distances of P and P ; there 
fore the law of the lever in all cases is the same, namely : 

The moment of the power equals the moment of the weight. 

When the forces act obliquely, the pressure on the fulcrum is 
less than the sum of the forces ; for, if C E is parallel to B D, 
then D E, E C, and C D, represent the three forces which are in 
equilibrium. But C D is less than the sum of D E and E C. 

110. The Compound Lever. When a lever acts on a 
second, that on a third, &c., the machine is called a compound 
lever. The law of equilibrium is 

TJie power is to the weight as the product of the acting distances 
on the side of the weight is to the product of the acting distances on 
the side of the power. 

Let the force exerted by A B on B D (Fig. 80) be called a;, and 

FIG. 80. 

x :\ B C:AC; 
y \\DF\BF\ 
W .:EG:DG. 

that of B D on D E be called y ; then 


and z 

Compounding these proportions, and dividing the first couplet by 
the common factors, we have 

P: W:: B C x D F x E G \ A C x B Fx D G. 

If the levers were of irregular forms, the acting distances might 
not be identical with the arms, as they are in the figure. 



111. The Balance. This is a common and valuable instru 
ment for weighing. It is a straight lever with equal arms, having 
scale-pans, either suspended at the ends, or standing upon them, 
one to contain the poises, and the other the substance to be 
weighed. For scientific purposes, particularly for chemical analy 
sis, great care is bestowed on the construction of the balance. 

The arms of the balance, measured from the fulcrum to the 
points of suspension, must be precisely equal. 

The knife-edges forming the fulcrum, and the points of sus 
pension, are made of hardened steel, and arranged exactly in a 
straight line. 

The centre of gravity of the beam is Mow the fulcrum, so that 
there may be a stable equilibrium ; and yet below it by an exceed 
ingly small distance, in order that the balance may be very 

To preserve the edge of the fulcrum from injury, the beam is 
raised by supports called Y s, when not in use. 

A long index at right angles to the beam, points to zero on a 
scale when the beam is horizontal. 

To protect the instrument from dust and moisture at all times, 
and from air-currents while weighing, the balance is in a glass 
case, whose front can be raised or lowered at pleasure. 

FIG. 81. 

A balance for chemical analysis is shown in Fig. 81. By turn 
ing the knob 0, the beam can be raised on the Y s A A from the 
surface on which the fulcrum K rests. The screw C raises and 


lowers the fulcrum in relation to the centre of gravity of the beam, 
in order to increase or diminish the sensitiveness of the instru 
ment. In the most carefully made balances, the index will make 
a perceptible change, by adding to the scale one millionth of the 

For commercial purposes, it is convenient to have the scale- 
pans above the beam. This is done by the use of additional bars, 
which with the beam form parallelograms, whose upright sides 
are rods, projecting upward and supporting the scales. Such con 
trivances necessarily increase friction ; but balances so constructed 
are sufficiently sensitive for ordinary weighing. 

112. The Steelyard. This is a weighing instrument, having 
a graduated arm, along which a poise may be moved, in order to 
balance various weights on the short arm. While the moment of 
the article weighed is changed by increasing or diminishing its 
quantity, that of the poise is changed by altering its acting dis 
tance. Since P x A O = W x B C (Fig. 82), and P is constant, 


FIG. 82. 

tic ~ba 

I" T"l " I I I 


and also the distance B C constant, A O cc TF; hence, if IF is suc 
cessively 1 lb., 2 Ibs., 3 Ibs., &c., the distances of the notches, a, b, c, 
&c., are as 1, 2, 3, &c. ; in other words, the bar C D is divided into 
equal parts. In this case, the graduation begins from the fulcrum 
C as the zero point. 

But suppose, what is often true, that the centre of gravity of 
the steelyard is on the long arm, and that P placed at E would 
balance it ; then the moment of the instrument itself is on the 
side C D, and equals P x C E. Hence, the equation becomes 

P x A C + P x CE = W x B C\ or 
P x A E = W x B C. 

.-. TT oc A E ; and the graduation must be considered as com 
mencing at E for the zero point. Such a steelyard cannot weigh 
below a certain limit, corresponding to the first notch a. 

To find the length of the divisions on the bar, divide A E, the 
distance of the poise from the zero point, by JF, the number of 
units balanced by P 9 when at that distance. 


The steelyard often has tivo fulcrums, one for less and the other 
for greater weights. 

113. Platform Scales. This name is given to machines 
arranged for weighing heavy and bulky articles of merchandise. 
The largest, for cattle, loaded wagons, &c., are constructed with 
the platform at the surface of the ground. In order that the plat 
form may stand firmly beneath its load, it rests by four feet on as 
many levers of the second order, whose arms have equal ratios. 
A F, B F, C G, D G (Fig. 83), are four such levers, resting on the 

FIG. 83. 

fulcrums, A, B, C, D, while the other ends meet on the knife- 
edge, F G, of another lever, L M. This fifth lever has its fulcrum 
at L, and its outer extremity is attached by a vertical rod, M N, 
to a steelyard, whose fulcrum is E, and poise P. The five levers 
are arranged in a square cavity just below the surface of the 
ground. The dotted line shows the outline of the cavity. On the 
bearing points of the four levers, H, /, J, /i, rest the feet of the 
platform (not represented), which is firmly built of plank, and just 
fits into the top of the cavity without touching the sides. The 
machine is a compound lever of three parts ; for the four levers 
act as one at F G, and are used to give steadiness to the platform 
which rests upon them. 

A construction quite similar to the above is made of portable 
size, and used in all mercantile establishments for weighing heavy 

114. Questions on the Lever. 1. A B (Fig. 84) is a uni 
form bar, 2 feet long, and weighs 
4 oz. ; where must the fulcrum be 

put, that the bar may be balanced ^ ^ 

by P, weighing 5 Ibs. ? 

Ans. 4 of an inch from A. 

T* Xlfrta 

2. A lever of the second order w 

is 25 feet long ; at what distance from the fulcrum must a weight 

FIG. 84. 


of 125 pounds be placed, so that it may be supported by a power 
able to sustain CO pounds, acting at the extremity of the lever. 

Ans. 12 feet. 

3. A and B are of the same height, and sustain upon their 
shoulders a weight of 150 pounds, placed on a pole 9} feet long; 
the weight is placed 6^ feet from A ; what is the weight sustained 
by each person ? 

Ans. A sustains 42 Ibs., and B sustains 107 \ Ibs. 

4. The longer arm of a steelyard is 2 feet 2 inches in length, 
and the shorter 2j inches; and its apparatus of hooks, &c., is so 
contrived that a weight of 2 pounds, placed upon the longer arm, 
at the distance of 10 inches from the centre of motion, will balance 
8 pounds placed at the extremity of the shorter arm ; the movable 
weight (of 2 pounds) cannot conveniently be placed nearer to the 
fulcrum than ~ of an inch; what must be the graduation of the 
steelyard that it may weigh ounces, and what will be the greatest 
and least weights that can be ascertained by it ? 

Ans. The graduation is to 12thsof an inch; and it will 
weigh from 1 to 20 pounds. 

5. A lent lever, A C B (Fig. 85), has the arm A G = 3 feet, 
B = 8 feet, P = 5 Ibs., and the an 
gle A C B = 140 ; what weight, W, 

must be attached at B, in order to 
keep A C horizontal ? 

Ans. 2.4476 Ibs. 

6. A cylindrical straight lever is 
14 feet long, and weighs 6 Ibs. 5 oz. ; 
its longer arm is 9, and its shorter 5 
feet ; at the extremity of its shorter 
arm a weight of 15 Ibs. 2 oz. is sus 
pended ; what weight must be placed 
at the extremity of the longer arm to 

keep it in equilibrium ? Ans. 7 Ibs. 

7. A uniform bar, 12 feet long, weighs 7 Ibs. ; a weight of 10 
Ibs. hangs on one end, and 2 feet from it is applied an upward 
force of 25 Ibs. ; where must the fulcrum be put to produce equi 
librium? Ans. 1 foot from the 10 Ibs. 

8. The lengths of the arms of a balance are a and I. When p 
ounces are hung on #, they balance a certain body ; but it requires 
q ounces to balance the same body, when placed in the other scale. 
What is the true weight of the body ? According to the first 
weighing, a p I x ; according to the second, I q a x. :. a bp q 
= a b x*, and x = ^jTq. Hence, the true weight is a geometrical 
mean between the apparent weights. 


9. On one arm of a false balance a body weighs 11 Ibs. ; on the 
other, 17 Ibs. 3 oz. ; what is the true weight ? 

Ans. 13 Ibs. 12 oz. 

10. Four weights of 1, 3, 5, 7 Ibs., respectively, are suspended 
from points of a straight lever, eight inches apart; how far from 
the point of suspension of the first weight must the fulcrum be 
placed, that the weights may be in equilibrium ? 

Ans. 17 inches. 

11. Two weights keep a horizontal lever at rest, the pressure 
on the fulcrum being 10 Ibs., the difference of the weights 4 Ibs., 
and the difference of the lever arms 9 inches; what are the weights 
and their lever arms ? 

Ans. Weights, 7 Ibs. and 3 Ibs. ; arms, 6| in. and 15| in. 

FIG. 86. 


115. Description and Law of the Machine. The wheel 
and axle consists of a cylinder and a wheel, firmly united, and free 
to revolve on a common axis. The power acts at the circumfer 
ence of the wheel in the direction of a tangent, and the weight in 
the same manner, at the circumference of the cylinder or axle ; so 
that the acting distances are the radii at the two points of contact. 
As the system revolves, the radii successively take the place of 
acting distances, without altering at all the relation of the forces 
to each other. The wheel and axle is therefore a kind of endless 

Let W (Fig. 86) be the weight suspended from the axle, tend 
ing to revolve it on the line L M\ 
and P, the power acting on the 
wheel, tending to revolve the sys 
tem in the opposite direction. It 
is plain that the acting distances 
are the radius of the axle, and A C 
the radius of the wheel. In case of 
equilibrium, the moment of W 
equals the moment of P. Calling 
the radius of the axle r, and the 
radius of the wheel P, then W x r 
= P x P; or 


If, instead of the weight P, suspended on the wheel, the rope 
be drawn by any force in the direction P or P", it is still tangent 
to the circumference, and therefore its acting distance, C DOT C B, 
the same as before. In general, the law of equilibrium for this 
machine is, 



FIG. 87. 

TJie power is to the weight as the radius of the axle to the radius 
of the wheel 

If the rope on the wheel, being fastened at A (Fig. 87) is 
drawn by the side of the wheel, as A P , 
the acting distance of the power is dimin 
ished from G A to G E, and therefore its 
efficiency is diminished in the same ratio. 
Were the rope drawn away from the 
wheel, as A P", making an equal angle 
on the other side of A P, the same effect 
is produced, the acting distance now be 
coming OF. 

The radius of the wheel and the radius 
of the axle should each be reckoned from 
the axis of rotation to the centre of the 

rope ; that is, half of the thickness of the rope should be added to 
the radius of the circle on which it is coiled. Calling t the half 
thickness of the rope on the axle, and t that of the rope on the 
wheel, the proportion for equilibrium is P : W : : r + t : 
R + t . 

116. The Compound Wheel and Axle. When a train 
of wheels, like that in Fig. 88, is put in 
motion, those which communicate mo 
tion by the circumference are called 
driving wheels, as A and O; those which 
receive motion by the circumference are 
called driven wheels. And the law of 
equilibrium is, 

TJie power is to the weight as the pro 
duct of the radii of the driving wheels 
to the product of the radii of the driven 

The crank P Q is to be reckoned 
among driven wheels ; the axle E among driving wheels. 

Let the radius of B be called R ; of D, R ; of A, r ; of 0, r ; 
of E, r". Call the force exerted by A on B, x ; that of C on D, y. 

P:x::r : P Q-, 
x \y ::r : R; 
y :W::r" :R r ; 

.: P: W . .r x r x r" : P Q x R x R 1 . 
If the driving wheels are equal to each other, and also the 
driven wheels, and the number of each is n, then 
P: W: : r n : R 1 . 

FIG. 88. 


117. Direction and Rate of Revolution. When two 
wheels are geared together by teeth, they necessarily revolve in 
contrary directions. Hence, in a train of wheels, the alternate 
axles revolve the same way. 

The circumferences of two wheels which are in gear move with 
the same velocity ; hence the number of revolutions will be re 
ciprocally as the radii of the wheels. 

Since teeth which gear together are of the same size, the rela 
tive nun^ber of teeth is a measure of the relative circumferences, 
and therefore of the relative radii of the wheels. If the wheel A 
(Fig. 88) has 20 teeth, and B has 40, and again if C has 15, and 
D 45, then for every revolution of J?, A revolves twice, and for 
every revolution of D, C revolves three times. Therefore, six 
turns of the crank are necessary to give one revolution to the 
axle E. 

By cutting the teeth of wheels on a conical instead of a cylin 
drical surface, the axles may be placed at any angle with eacli 
other, as represented in Fig. 89. 

Whether axles are parallel or not, lands in- FIG- 89. 

stead of teeth may be used for transmitting 
rotary motion. But as bands are liable to slip 
more or less, they cannot be employed in cases 
requiring exact relations of velocity. 

118. Questions on the Wheel and 

1. A power of 12 Ibs. balances a weight of 
100 Ibs. by a wheel and axle; the radius of the 
axle is 6 inches ; what is the diameter of the wheel ? 

Ans. 8 ft. 4 in. 

2. W 500 Ibs. ; R = 4 ft. ; r = 8 in. ; the weight hangs by a 
rope 1 inch thick, but the power acts at the circumference of the 
wheel without a rope ; what power will sustain the weight ? 

Ans. 88.54 Ibs. 

3. R = 1 ft; r 2 in.; the well-stone weighs 256 Ibs.; the 
bucket, empty, weighs 18 Ibs. ; the bucket, filled, weighs 65 Ibs. ; 
what force must a person apply to the bucket-rope, in each case, 
for equilibrium? Ans. 1st, down, 24 Ibs.; 2d, up, 22| Ibs. 

4. In Fig. 88, A and C have each 15 teeth, B and D each 40 
teeth ; the radius of the axle E is 4 inches ; the rope on it 1 inch 
in diameter ; and the radius of the crank P Q is 18 inches ; what 
is the ratio of power to weight in equilibrium ? Ans. 1 : 28$. 



FIG. 90. 

FIG. 91. 


119. The Pulley Described. The pulley consists of one or 
more wheels or rollers, with a rope passing over the edge in which 
a groove is sunk to keep the rope in place. The axis of the roller 
ir? in a Nock, which is sometimes fixed, and sometimes rises and 
foils with the weight; and the pulley is accordingly called a fixed 
pulley or a movable pulley. The principle which explains the 
relation of power and weight in every form of pulley, is this : 

Whatever strain or tension is applied to one end of a cord, is 
transmitted through its whole length, if it does not branch, however 
much its direction is changed. 

In the pulley, the sustaining portions of the rope are assumed 
to be parallel to each other. 

120. The Fixed Pulley. In 

the fixed pulley, A (Fig. 90), the 
force P produces a tension in the 
string, which is transmitted through 
its whole length, and which can be 
balanced only when W equals P. 
Hence, in the fixed pulley, the 
power and weight are equal This 
machine is useful for changing the 
direction in which the force is ap 
plied to the weight; and if the 
power only acts in the plane of the 
groove of the wheel, it is immaterial what is its direction, horizon 
tal, vertical, or oblique. 

121. The Movable Pulley. In Fig. 91, the ten 
sion produced by P, is transmitted from A down to 
the wheel E, and thence up to D ; therefore W is sus 
tained by two portions of the rope, each of which 
exerts a force equal to P. 

:. W= 2P; or P: TF::1 : 2. 

The same reasoning applies, where the rope passes 
between the upper and lower blocks any number of 
times, as in Fig. 92. The force causes a tension in the 
rope, which is transmitted to every portion of it. If n 
is the number of portions which sustain the lower 
block, then W is upheld by n P ; and if there is equi 
librium, P : W : : 1 : n. In the figure, the weight 
equals six times the power. The law of equilibrium, 
therefore, for the movable pulley with one rope, is this, 

TJie power is to the weight as one to the number of 

FIG. 92. 

ii".:. ." < li"^ 1 " -.! "> 



the sustaining portions of the rope; or, as one to twice the number 
of movable pulleys. 

122. The Compound Pulley. Wherever a system of pul 
leys has separate ropes, the machine is to be regarded as com 
pound, and its efficiency 

is calculated according- ^ 93 - *"" 94 - 

ly. Figures 93 and 94 
are examples. In Fig. 
93, call the weight sus 
tained by F, x, and that 
sustained by D, y. Then 
(Art. 1^1), 


And if n is the number 
of ropes, 

P : W : : 1 : 2 n . 

In Fig. 94, the tension P is transmitted 
over A directly to the weight at G ; the wheel 
A is loaded, therefore, with 2 P, and a tension 
of 2 P comes upon the second rope, which is 
transmitted over B to the weight at F. In 
like manner, a tension of 4 P is transmitted 
over to E. The sum of all these being ap 
plied to the weight, it must therefore be equal 
to that sum in case of equilibrium. Therefore, P : W : : 1 : 1 -f- 
2 + 4 + &c. Now the sum of this geometrical series to n terms 
is Z n 1 ; /. P : W: : 1 : 2 n 1. This combination is therefore a 
little less efficient than the preceding. 

Since the several ropes have different tensions, the weight can 
not be balanced upon them, unless those of greatest tension are 
nearest the line of direction of the body. For example, if the rope 
Pis directed toward the centre of gravity of the. weight, the rope 
G should be attached four times as far from it as the rope E, in 
order to prevent the weight from tipping. 

The pulley owes its efficiency as a machine to the fact, that the 
tension produced by the power is applied repeatedly to the weight. 
The only use of the wheels is to diminish friction. Were it not 
for friction, the rope might pass round fixed pins in the blocks, 
and the ratio of power to weight would still be in every case the 
same as has been shown. 



FIG. 95. 


123. Definition and Law of this Machine. 

Tfie rope machine is one in which the power and weight are in, 
equilibrium ly the tension of one or more ropes. 

According to this definition the pulley is included. It is that 
particular form of the rope machine in which the sustaining parts 
of the ropes are parallel; and it is treated as a separate machine, 
because its theory is very simple, and because it is used far more 
extensively than any other forms. 

If the two portions of rope 
which sustain the weight are in 
clined, as in Fig. 95, then W is 
no longer equal to the sum of 
their tensions, as it is in the pul 
ley, but is always less than that, 
according to the following law : 

TJie power is to the weight as 
radius is to twice the cosine of 
half the angle between the parts 
of the rope. 

Put A E B = 2 a ; then 
FED = a, and since sin B E W 

= sin B ED sin a, we shall have (Art. 61) P : W: sin a 
2 a ; but sin a : sin 2 a : : R : 2 cos a ; .. P : W : : R : 2 cos a. 

If in Fig. 96, the end of the cord, instead of being attached to 
the beam, is carried over another fixed pulley, and a weight equal 
to P is hung upon it, the equilibrium will be preserved, because all 
parts of the rope have a tension equal to P; therefore, as before, 


124. Change in the Ratio of Power and Weight. If 

P is given, all the possible values of W are included between 
IT =0, and W=2P. 

When the rope is straight from A to B, so that C D 0, then, 
by the above proportion, W = 0. As W is increased from zero, 
the point C descends; and 
when D C = $ B C, then, by 
the proportion, W = P. In 
that case D C B = 60, and 
the angles, A C B, A C W, 

FlQ - 9( 5. 

and B C JF, are equal (each 
being 120), as they should be, 
because each of the equal 
forces, P, P, and W, is as the 
sine of the angle between the directions of the other two. 



But when W has increased to 2 P, it descends to an infinite 
distance; for then, by the proportion, CD = B G, that is, the side 
of a right-angled triangle is equal, to the hypothenuse. Thus, the 
extreme values of W are and 2 P. 

It appears from the foregoing, that a perfectly flexible rope 
having weight cannot be drawn into a straight horizontal line, by 
any force however great ; for G cannot coincide with /), except 
when W - 0. 

125. The Branching Rope. When (7, where the weight is 
suspended, is a fixed point of the rope, 

we have a branching rope, and the 
principle of transmitted tension does 
not apply beyond the point of division. 
Let P, > and W (Fig. 97), be 
given, and C a fixed point of the rope. 
Produce W C, and let A E, drawn 
parallel to G B, intersect it in E. The 
sides of A G E are proportional to the 
given forces ; therefore its angles can 
be found, and the inclinations of A G 
and B G to the vertical G W are known. 

126. The Funicular Polygon. If several weights are at 
tached at fixed points along the cord A G B (Fig. 98), the combi 
nation is called the funicular polygon ; 

and the fact that there are opposite 
and equal tensions in any portion of 
the cord will enable us to transfer all 
the forces to one point. 

Let the tension of G D T\ and 
that of D E = T . C is kept at rest 
by P, T, and W; hence T is equal to 

the resultant of P and W. But the same T (in the opposite di 
rection) equals and balances the resultant of T and W . Suppose, 
now, C D to vanish, by removing G to D ; draw A D parallel to 
A G-, let P act in the line D A, and Win the line D W. Dmll 
now be in equilibrium, as before, because there has merely been made 
a substitution of P and W in their original directions for T, their 
equivalent. Now consider D to be acted on by three forces, T , P, 
and W + W; . . T = the resultant of P and W + W , and the 
two latter can be transferred, as before, to E,AE being parallel to 
A C or A D. E is therefore kept at rest by the three forces, P, P , 
and W -f W + W". We can now use the triangle of forces, as in 
the preceding article, to determine the directions of B E and A E, 
or its parallel, A C, and hence, of the parts, G D and D E, 

FIG. 98. 



If the weights are all equal, and their number = ft, the three 
forces at E are P, F , and n IF. 

An example of this kind occurs in 
the suspension bridge, whose weight is 
distributed at equal distances along 
the supporting chains. And an ex 
treme case is that of a heavy rope or 
chain suspended loosely over pulleys, 
as in Fig. 99. Equal weights are sus 
pended at an infinite number of points, 
and therefore the funicular polygon becomes a curve, and is called 
the catenary curve. Its directions at the extremities, A and B, 
and the law of the curve, may be determined by the principles 
given above. 


127. Relation of Power, "Weight, and Pressure on the 
Plane. The mechanical efficiency of the inclined plane is ex 
plained on the principle of oblique action ; that is, it enables us to 
apply the power to balance or overcome only one component of the 
weight, instead of the whole. Let the weight of the body G, lying 
on the inclined plane A C (Fig. 100), be represented by IF; and 
resolve it into F parallel, and N perpendicular to the plane. N 
represents the perpendicular pressure, and is equal to the reaction 
of the plane ; F is the force by which the body tends to move 
down the plane. 

Let a = the angle C, the inclination of the plane ; therefore 
WG tf = a. Then F= fT.sma; and^V= JF.cosa. 

FIG. 100. 

FIG. 101. 

Now suppose a force P is applied at (Fig. 101), which keeps 
the body at rest. Then the resultant of W and P must be N 9 
which is resisted by the plane ; therefore, 

P : W: : sin G N P, or sin a : sin P G N. 

When the power acts parallel to the plane, P G N = 90, and 
we have P : W : : sin a : sin 90 : : A B : A C. Hence, when the 


power acts in a line parallel to the inclined plane, which is the 
most common direction, 

The power is to the iveight as the height to the length of the in- 
dined plane. 

When the power acts in a line parallel to the base of the in 
clined plane, P G N = 90 a, and we have P : W : : sin a : 
cos a : : A B : B C. Hence, when the power acts in a line parallel 
to the base of the inclined plane, 

The power is to the weight as the height is to the lase of the in 
clined plane. 

128. Power most Efficient when Acting Parallel to 
the Plane. From the proportion 

P : W : : sin a : sin P G N, we derive 
P.sinP GN 

sin a 

Now as P and sin a are given, W varies as sin P G N, which 
is the greatest possible when P G N = 90 ; that is, when the 
power acts in a line parallel to the plane. 

Whether the angle P G N diminishes or increases from 90, 
its sine diminishes, and becomes zero, when P G N= 0, or 180. 
Therefore W 0, or no weight can be sustained, when the power 
acts in the line G- jV", perpendicular to the plane, either toward the 
plane or from it. 

129. Expression for Perpendicular Pressure. From 
the triangle P G N^NQ obtain 

N: W: : sin P N: sin P G N, 
or N: W: : sin P G W: sin P G N 

_ w 


-,-fc _^>j -T- 

sm P G N 
If the power acts in a line parallel to the inclined plane, 

P 0TF=90 + oP0 J y=90ana JV = W (Si 

iii y \j 
W cos a. 

If the power acts in a line parallel to the base of the inclined 

plane, P G W = 90, P G N = 90 - a, and N = W = 

cos a 

W sec a. 

If the power acts in a line perpendicular to the inclined plane, 

P G W = a, P G N = 0, and N = W -~ - oc. 

130. Equilibrium between Two Inclined Planes. If a 

body rests, as represented in Fig. 102, between two inclined planes, 



FIG. 103. 

the three forces which retain it are its weight, and the resistances 
of the planes. Draw H F and L F 
perpendicular to the planes through 
the points of contact, and G F verti 
cally through the centre of gravity of 
the body. Since the body is in equi 
librium, these three lines will pass 
through the same point (Art. 60, 2). 
Let that point be F, and draw G P 
parallel to L F, and M K parallel to 
the horizon. GPFis similar to K C M. 
Therefore (since Pressure on A C : Pr. 

Pressure on A G : Pr. on D C 

: sin M : sin K, 

:sinl> C E : sin A C B. 

FIG. 103. 

That is, when a body rests between two planes, it exerts pressures 
on them which are inversely as the sines of their inclinations to 
the horizon. 

If, therefore, one of the planes is horizontal, none of the 
pressure can be exerted on any other plane. It is friction alone 
which renders it possible for a body on a horizontal surface to 
lean against a vertical wall. 

131. Bodies Balanced on Two Planes by a Cord 
passing over the Ridge. Let P and W balance each other on 
the planes A D and A C (Fig. 103), 
which have the common height A B, 
by means of a cord passing over the 
fixed pulley A. The tension of the 
cord is the common power which pre 
vents each body from descending ; and 
as the cord is parallel to each plane, 
we have (calling the tension t), 


that is, the weights, in case of equilibrium, are directly as the 
lengths of the planes. 

132. Questions on the Inclined Plane. 

1. If a horse is able to raise a weight of 440 Ibs. perpendicu 
larly, what weight can he raise on a railway having a slope of five 
degrees? Am. 5048-5 Ibs. 


t : W 
P-. W 

AD: AC; 


2. The grade of a railroad is 20 feet in a mile ; what power 
must be exerted to sustain any given weight upon it ? 

Ans. 1 Ib. for every 264 Ibs. 

3. "What force is requisite to hold a body on an inclined plane, 
by pressing perpendicularly against the plane ? 

Ans. An infinite force. 

4. A certain power was able to sustain 500 tons on a plane of 
7i ; but on another plane, it could sustain only 400 tons ; what 
was the inclination of the latter ? Ans. 9 23 25". 

5. Equilibrium on an inclined plane is produced when the 
power, weight, and perpendicular pressure are, respectively, 9, 13, 
and 6 Ibs. ; what is the inclination of the plane, and what angle 
does the power make with the plane ? 

Ans. a == 37 21 26". Inclination of power to plane 
= 28 46 54". 

6. A power of 10 Ibs., acting parallel to the plane, supports a 
certain weight ; but it requires a power of 12 Ibs. parallel to the 
base to support it. What is the weight of the body, and what is 
the inclination of the plane ? 

Ans. W = 18.09 Ibs. a = 33 33 ; 25" 

7. To support a weight of 500 Ibs. upon an inclined plane of 
50 inclination to the horizon, a force is applied whose direction 
makes an angle of 75 with the horizon. What is the magnitude 
of this force, and the pressure of the weight against the plane ? 

Ans. P = 422.6 Ibs. N= 142.8 Ibs. 


133. Reducible to the Inclined Plane. The screw is a 
cylinder having a spiral ridge or thread around it, which cuts at a 
constant oblique angle all the lines of the surface parallel to the 
axis of the cylinder. A hollow cylinder, called a nut, having a 
similar spiral within it, is fitted to move freely upon the thread of 
the solid cylinder. In Fig. 104, let the base A B of the inclined 

FIG. 104. 


plane A C be equal to twice the circumference of the cylinder 
A 1 E\ then let the plane be wrapped about the cylinder, bringing 
the points A, F, and B, to the point A -, then will A G describe 
two revolutions of the thread from A to C . Therefore the me 
chanical relations of the screw are the same as of the inclined 

If a weight be laid on the thread of the screw, and a force be 
applied to it horizontally in the direction of a tangent to the 
cylinder, the case is exactly analogous to that of a body moved on 
an inclined plane by a force parallel to the base. Let r be the 
radius of the cylinder, then 2 ?r r is the circumference ; also let d 
be the distance between the threads, (that is, from any point of 
one revolution to the corresponding point of the next,) measured 
parallel to the axis of the cylinder ; then 2 TT r is the base of an 
inclined plane, and d its height. Therefore (Art. 127), 
P: WndiZirr; or, 

TJie power is to the weight as the distance between the threads 
measured parallel to the axis, is to the circumference of the screw. 

If instead of moving the weight on the thread of the screw, the 
force is employed to turn the screw itself, while the weight is free 
to move in a vertical direction, the law is the same. Thus, 
whether the screw A 1 E is allowed to rise and fall in the fixed nut 
G Hj or whether the nut rises and falls on the thread of the screw, 
while the latter is revolved, without moving longitudinally, in 
each case, P : W : : d : 2 TT r. 

134. The Screw and Lever Combined. The screw is so 
generally combined with the lever in practical mechanics, that it 
is important to present the law of the 
compound machine. Let A F (Fig. 105) Frc^lOS. 

be the section of a screw, and suppose 
B C, a lever of the second order, to be 
applied to turn it. The fulcrum is at C, 
the power acts at B, and the effect pro 
duced by the lever is at A, the surface of 
the cylinder. Call that effect x, and let 
d = the distance between the threads ; 

P: x .-.AC-.BC, 
and xi W:: d:2nAC , 
compounding and reducing, we have 

P:W:i d .lnBC , that is, 

TJie power is to the weight as the distance between the threads, 
measured parallel to the axis, to the circumference described by the 



FIG. 106. 

The law as thus stated is applicable to the screw when used 
with the lever or without it. 

135. The Endless Screw. The screw is so called, when its 
thread moves between the teeth of a wheel, thus causing it to 
revolve. It is much used for diminish 
ing very greatly the velocity of the 

Let P Q (Fig. 106) be the radius of 
the crank to which the power is ap 
plied; d, the distance between the 
threads; R, the radius of the wheel; 
r, the radius of the axle ; and call the 
force exerted by the thread upon the 
teeth, x. Then, 

x : : d : 2 TT x P Q, 
W:: r :E; 

?nd x 

.-. P 

W ::dr:2TT x R x P Q. 

If, for example, P Q = 30 inches, d = 1 in., R = 18 in. ; 
r + t = 2 in. ; then W moves with 1696 times less velocity than P. 

136. The Right and Left Hand Screw. The common 
form of screw is called the right-hand screw, and may be described 
thus ; if the thread in its progress along the length of the cylinder, 
passes from the left over to the right, it is called a right-hand screiv. 
Hence, a person in driving a screw forward turns it from his left 
over (not under) to his right, and in drawing it back he reverses 
this movement. Fig. 104 represents a right-hand screw. 

The left-hand screw is one whose thread is coiled in the oppo 
site direction, that is, it advances by passing from right over to 
left. This kind is used only when there is special reason for it. 
For example, the screws which are cut upon the left-hand ends of 
carriage axles are left-hand screws; otherwise there would be 
danger that the friction of the hub against the nut might turn 
the nut off from the axle. Also, when two pipes for conveying 
gas or steam are to be drawn together by a nut, one must have a 
right-hand, and the other a left-hand screw. 

137. Questions on the Screw. 

1. The distance between the threads of a screw is one inch, the 
bar is two feet long from the axis, and the power is 30 Ibs. ; what 
is the weight or pressure ? Ans. 4523.76 Ibs. 

2. The bar is three feet long, reckoned from the axis, P 
60 Ibs., W = 2240 Ibs. ; what is the distance between the threads ? 

Ans. 6.058 inches. 



3. A compound machine consists of a crank, an endless screw, 
a wheel and axle, a pulley, and an inclined plane. The radius of 
the crank is 18 inches ; the distance between the threads of the 
screw, one inch ; the radius of the wheel on which the screw acts, 
two feet ; the radius of the axle, G inches ; the pulley block has two 
movable pulleys with one rope ; and the inclination of the plane to 
the horizon is 30. What weight on the plane will be balanced by 
a power of 100 Ibs. applied to the crank ? Am. 361911.108 Ibs. 


138. Definition of the Wedge, and the Mode of 
Using. The usual form of the wedge is a triangular prism, two 
of whose sides meet at a very acute angle. This machine is used 
to raise a weight by being driven as an inclined plane underneath 
it, or to separate the parts of a body by being driven between 
them. When it is used by itself, and does not form part of a 
compound machine, force is usually applied by a blow, which pro 
duces an intense pressure for a short time, sufficient to overcome 
a great resistance. 

139. Law of Equilibrium. Whatever be the direction of 
the blow or force, we may suppose it to be resolved into two com 
ponents, one perpendicular to the back of the wedge, and the other 
parallel to it. The latter produces no effect. The same is true of 
the resistances ; we need to consider only those components of them 
which are perpendicular to the sides of the wedge. 

Let M N (Fig. 107) represent a section of 
the wedge perpendicular to its faces ; then P A, 
Q A y and R A, drawn perpendicular to the faces 
severally, show the directions of the forces which 
hold the wedge in equilibrium. Taking A B to 
represent the power, draw B C parallel to R A, 
and we have the triangle A B C, whose sides 
represent these forces. But A B C is similar to 
N N 0, as their sides are respectively perpen 
dicular to each other. Hence, calling the forces 
P, Q, and R, respectively, 

P: QnMNiMO; 

andP:7Z:: M N: NO; 

that is, there is equilibrium in a wedge, when 

The power is to the resistances as the back of 
the wedge to the sides on which the resistances respectively act. 

If the triangle is isosceles, the two resistances are equal, as the 
proportions show ; and P is to either resistance, R, as the breadtli 
of the back to the length of the side. 


If tlie resisting surfaces touch the sides of the wedge only in 
one point each, then Q A and R A, drawn through the points of 
contact, must meet A P in the same point (Art. 60, 2) ; otherwise 
the wedge will roll, till one face rests against the resisting body 
in two or more points. 

The efficiency of the wedge is usually very much increased by 
combining its own action with that of the lever, since the point 
where it acts generally lies at a distance from the point where the 
effect is to be produced. Thus, in splitting a log of wood, the re 
sistance to be overcome is the cohesion of the fibers ; and this force 
is exerted at a distance from the wedge, while the fulcrum is a 
little further forward in the solid wood. 


140. Description and Law of Equilibrium. The knee- 
joint consists of two bars, usually equal, hinged together at one 
end, while the others are at liberty to separate in a straight line. 
The power is applied at the hinge, tending to thrust the bars into 
a straight line ; the weight is the force which opposes the separa 

FIG. 108, 

Suppose that A B and A D (Fig. 108) are equal bars, hinged 
together at A ; and that the bar A B is free only to revolve about 
the axis B, while the end D of the other bar can move parallel to 
the base E F. HP urges A toward the base, it tends to move D 
further from the fixed point B. The force P , which opposes that 
motion, is represented in the figure by the weight W. The law of 
equilibrium is, 

The power is to the weight as twice radius to the tangent of half 
the angle between the bars. 

The point A is held in equilibrium by three forces, the power 
P, the resistance along B A, and that along D A. As A D is 


isosceles, and A C is perpendicular to B D, the angles P A B and 
PAD are equal ; therefore the resistances B A and D A are 
equal (Art. 61). Let a = the angle B A C - B A D\ and let 
R = the resistance in the line D A. Then, since sin P A B=sin a, 
we have 

P : R : : sin 2 a : sin a. 

But TF is equal only to that component of R which is parallel to 
B D. Therefore, resolving E, we have (Art. 49), 

R : W: : rad : cos A D G\ or 
: : rad : sin a. 

Compounding, we find, 

P : W : : rad . sin 2 a : sin 8 a. 

2 sin a . cos a 

But sin 2 a = - : therefore 


D rrr 2 rad . sin a . cos a 

P : W - - : sm a a ; 


, rad . sin a 

: : 2 rad : = tan a 

cos a 

or the power is to the weight as twice radius to the tangent of the 
half angle between the bars. 

Since A C : CD:: rad : tan a, 

.-. P:W::2A C: CD; or, 

The power is to the weight as twice the height of the joint to 
half the distance between the ends of the bars. 

141. Ratio of Power and Weight Variable. It is obvi 
ous that the ratio between power and weight is different for differ 
ent positions of the bars. As A is raised higher, BAD diminishes ; 
and when B A D = 0, then a = 0, and tan a = ; 

/. P: TF::2rad:0, 

and the power has no efficiency. But as A approaches the base 
E F, a approaches 90 ; therefore tan a increases, and the power is 
more efficient. When A, B, and D are in a straight line, a = 90, 
and tan a is infinite, .*. P : W : : 2 rad : <x . Hence, the efficiency 
of the power is infinitely great. The indefinite increase of effi 
ciency in the power, which occurs during a single movement, ren 
ders this machine one of the most useful for many purposes, as 
printing and coining. 

Questions on the knee-joint. 

1. A power of 50 Ibs. is exerted on the joint A (Fig. 108) ; 
compare the weight which will balance it, when B A D is 90, 
and when it is 160. Ans. 25 Ibs. and 141.78 Ibs. 

2. When the angle between the bars is 110, a certain power 


just overcomes a weight of 65 Ibs. ; what must be the angle, in 
order that the weight overcome may be five times as great ? 

Am. 164 3 22". 


142. The Point of Application Moving in the Line of 
the Force. In examining the simple machines, we have in each 
instance simply inquired for the relative magnitude of the forces, 
called the power and the weight, when in equilibrium. There is 
another important particular to be noticed, namely, the relative 
velocity of the power and weight, when they begin to move. It 
can be shown, in every case, that the velocities, when reckoned in 
the direction in which the forces act, are inversely as the forces. 

Some examples are first given in which the point of application 
moves in the line in which the force acts. 

In the straight lever (Fig. 109), which is in equilibrium by the 
weights P and W, suppose a 

slight motion to exist ; then FIG. 109. 

the velocity of each will be 
as the arc described in the 
same time ; but the arcs are 
similar, since they subtend 
equal angles. Therefore, if V = velocity of P, and v = velocity 

of IF, 

V: v:\AP:B W::A C:B 0; 

but it has been shown (Art. 106) that 

P-.W . .B C:A C-, 
.\V:v:: W : P , 

that is, the velocity of the power is to the velocity of the weight 
as the weight to the power. Hence, P x its velocity = W x its 
velocity ; that is, the momentum of the power equals the mo 
mentum of the weight. 

In the wheel and axle, let R and r be the radii, and suppose the 
machine to be revolved ; then while P descends a distance equal 
to the circumference of the wheel = 2 TT R, the weight ascends a 
distance equal to the circumference of the axle = 2 n r. There 

V:v\ :2n lt:2nr:: R .r; 
but (Art. 115), P : W: : r : R-, 

/. V: v :: W:P; 

or, the velocities are inversely as the weights ; and P x V = W x v, 
the momentum of the power equals the momentum of the weight- 
In the fixed pulley the velocities are obviously equal ; and we 
have before seen that the power and weight are equal ; therefore 



the proportion holds true, V: v : : W: P\ and the momenta are 

In the movable pulley, if n is the number of sustaining parts 
of the cord, when W rises any distance .T, each portion of cord is 
shortened by the distance x, and all these n portions pass over to 
P 9 which therefore descends a distance = n x. 

Hence, V : v : : n x : x : : n : 1 ; 

but (Art. 121), P: W:: 1 :M; 

/. V:v. .W .P] 
as in all the preceding cases. 

In the screw (Fig. 104), while the power describes the circum 
ference = 2 TT x B C, the weight moves only the distance = d ; 

V: v: : 2 n x B C: d; 
but (Art. 133), P : W : : d : 2 n x B C; 

;. V .v:-. W\ P; 

therefore the momentum of the power equals the momentum of 
the weight, as before. 

143. The Point of Application Moving in a Different 
Line from that in which the Force Acts. The cases thus 
far noticed are the most obvious ones, because the points of appli 
cation of power and weight actually move in the directions in 
which their force is exerted. But the principle we are considering 
is that of virtual velocities. If the force is exerted in one line, and 
the motion of the point of application is in a different line, then 
its virtual velocity is merely that component which lies in the 
former line. The case of the inclined plane will illustrate the 

First, let P (Fig. 110) act parallel to the plane, and suppose 
the body to be moved either up or down the plane a distance equal 
to G d. That is the velocity 
of the power. But in the di 
rection of the weight (force of 
gravity), the body moves only 
the distance I d. Therefore 
the velocity of the power is to 
the velocity of the weight 
(each being reckoned in the 
line of its action) as G d to 

By similar triangles, G d:b d:: A C: A B\ 
or V: v ::A C : A B. 

But (Art. 127), P: W \\AB\AC\ 

.-. V: v :: W : P. 

FIG. 110. 


Again, let the power act in any oblique direction, as G e. If 
the body moves over G d, draw d e perpendicular to G e ; then G e 
is the distance passed over in the direction of the power, and ~b d in 
the direction of the weight. G d being taken as radius, G e is 
cos d G e ^= cos (P G N 90) = sin P G N\ and I d = sin a. 
Therefore, the virtual velocity of the power is to the virtual velo 
city of the weight as sin P G N to sin a 

or V : v : sin P G N : sin a. 

But (Art. 128), P : W: 

.-. F: v : 

sin a : sin P G 

We learn from the foregoing principle, that a machine does not 
enable us to obtain any greater effect than the power could pro 
duce without its aid, but only to produce an effect in a different 
form. A given power, for instance, may move a much greater 
quantity of matter by the aid of a machine, but it will move it as 
much more slowly. On the other hand, a power, by means of a 
machine, may produce a far greater velocity than would be possible 
without such aid ; but the quantity moved, or the intensity of the 
force exerted, would be proportionally less. By machines, there 
fore, we do not increase the effects of a power, but only modify 


144. The Power and Weight not the only Forces in 
a Machine. For each machine a certain proportion has been 
given, which ensures equilibrium. And it is implied that if either 
the power or the weight be altered, the equilibrium will be de 
stroyed. But practically this is not true ; the power *or weight 
may be considerably changed, or possibly one of them may be en 
tirely removed, and the machine still remain at rest. The obstruc 
tion which prevents motion in such cases, and which always exists 
in a greater or less degree, arises from friction ; and friction is 
caused by roughness in the surfaces which rub against each other. 
The minute elevations of one surface fall in between those of the 
other, and directly interfere with the motion of either, while they 
remain in contact. Polishing diminishes the friction, but can 
never remove it, for it never removes all roughness. 

As friction always tends to prevent motion, and never to pro 
duce it, it is called a passive force. It assists the power, when the 
weight is to be kept at rest, but opposes it, when the weight is to 
be moved. There are other passive forces to be considered in the 
study of science, but no other has so much influence in the opera 
tions of machinery as friction. 

145. Modes of Experimenting. When one surface slides 
on another, the friction which exists is called the sliding friction ; 


but when a wheel rolls along a surface, the friction is called rolling 
friction. The sliding friction occurs much more in machines 
than the rolling friction. 

Experiments for ascertaining the laws of friction may be per 
formed by placing on a table a block of three different dimensions, 
and measuring its friction un 
der different circumstances by 
weights acting on the block by 
means of a cord and pulley, as 
represented in Fig. 111. This 
was the method by which Cou 
lomb first ascertained the laws 
of friction. 

Another mode is to place the block on an inclined plane, whose 
angle can be varied, and then find the relative friction in different 
cases, by the largest inclination at which it will prevent the block 
from sliding. For, when W on the 
inclined plane A B (Fig. 112), is on 
the point of sliding down, friction 
is the power, which acting parallel 
to the plane, is in equilibrium with 
the weight. In such cases, the 
power is to the weight as the height 
to the length. 

The coefficient of friction is the fraction whose numerator is 
the force required to overcome the friction, and its denominator 
the weight of the body. 

146, Laws of Sliding Friction. The laws of sliding fric 
tion on which experimenters are generally agreed are the fol 
lowing: * 

1. Friction varies as the pressure. If weights are put upon 
the block, it is found that a double weight requires a double force 
to move it, a triple weight a triple force, &c. 

2. It is the same, hoivever great or small the surface on which the 
body rests. If the block be drawn, first on its broadest side, then on 
the others in succession, the force required to overcome friction is 
found in each case to be the same. Extremes of size are, how 
ever, to be excepted. If the loaded block were to rest on three 
or four very small surfaces, the obstruction might be greatly 
increased by the indentations thus occasioned in the surface 
beneath them. 

3. Friction is a uniformly retarding force. That is, it destroys 
equal amounts of motion in equal times, whatever may be the 
velocity, like gravity on an ascending body. 


4. Friction at the first moment of contact is Jess than after con 
tact has continued for a time. And the time during which fric 
tion increases, varies in different materials. The friction of wood 
on wood reaches its maximum in three or four minutes ; of metal 
on metal, in a second or two ; of metal on wood, it increases for 
several days. 

5. Friction is less between substances of different Tcinds than 
between those of the same kind. Hence, in watches, steel pivots 
are made to revolve in sockets of brass or of jewels, rather than 
of steel. 

147. Friction of Axes. In machinery, the most common 
case of friction is that of an axis revolving in a hollow cylinder, or 
the* reverse, a hollow cylinder revolving on an axis. These are 
cases of sliding friction, in which the power that overcomes the 
friction, usually acts at the circumference of a wheel, and there 
fore at a mechanical advantage. Thus, the friction on an axis, 
whose coefficient is as high as 20 per cent., requires a power of 
only two per cent, to overcome it, provided the power acts at the 
circumference of a wheel whose diameter is ten times that of 
the axis. 

148. Rolling Friction. This form of friction is very much 
less than the sliding, since the projecting points of the surfaces do 
not directly encounter each other, but those of the rolling wheel 
are lifted up from among those of the other surface, as the wheel 

By the use of the apparatus described in Art. 145, the laws of 
the rolling are found to be the same as those of the sliding fric 
tion. But on account of the manner in which this form of fric 
tion is overcome, there is this additional law : f 

The force required to roll the wheel varies inversely as the 

For the power, acting at the centre of the wheel to turn it on 
its lowest point as a momentary fulcrum, has the advantage of 
greater acting distance as the diameter increases. 

It is the rolling friction which gives value to friction tuheels, 
as they are called. "When it is desirable that a wheel should 
revolve with the least possible friction, each end of its axis is made 
to rest in the angle between two other wheels placed side by side, 
as shown in Fig. 113. The wheel is obstructed only by the rolling 
friction on the surfaces of the four wheels, and the retarding effect 
of the sliding friction at the pivots of the latter is greatly reduced 
on the principle of the wheel and axle. 

The sliding friction is diminished by lubricating the surface, 
the rolling friction is not. 


FIG. 113. 

149. Advantages of Friction. Friction in machinery is 
generally regarded as an evil, since* more power is on this account 
required to do the work for which the machine is made. But it is 
easy to see, that in general friction is of incalculable value, or 
rather, that nothing could be accomplished without it. Objects 
stand firmly in their places by friction ; and the heavier they are, 
the more firmly they stand, because friction increases with the 
pressure. All fastening by nails, bolts, and screws, is due to fric 
tion. The fibers of cotton, wool or silk, when intertwined with 
each other, form strong threads or cords, only because of the power 
of friction. Without friction, it would be impossible to walk or 
even to stand, or to hold anything by grasping it with the hand. 



fl.50. The Force which Moves a Body Down an In 
clined Plane. It was shown (Art. 127) that when the power 
acts in a line parallel to the inclined plane, P :W:: A B : A C. 
If, therefore, P ceases to act, the body descends the plane only 
with a force equal to P. 

Let g (the velocity acquired in a second in falling freely) = the 
force of gravity, / = the force acting down the plane, li the 
height, I the length ; then by substitution, 


f:g::h:l, and 

Therefore, the force which moves a body down an inclined 
plane is equal to that fraction of gravity which is expressed by the 
height divided by the length. This is evidently a constant force 
on any given plane, and produces uniformly accelerated motion. 
Therefore the motion on an inclined plane- does not differ from 
that of free fall in kind, but only in degree. Hence the formulas 
for time, space, and velocity on an inclined plane are like those 
relating to free fall, if the value of/ be substituted for g. 

151. Formulas for the Inclined Plane. The formulae for 
free fall (Art. 28) are here repeated, and against them the corre 
sponding formulae for descent on an inclined plane. 

Free fall. Descent on an inclined plane. 

f-i/ . 

21 " 






o . 

1 ,^ 





- Va~l 

A/% 9 ]l * 

v i/ v 




- ^ 




By formula 1 , s <x j* , and by formula 3, s cc v 2 . It follows that 
in equal successive times the spaces of descent are as the odd 
numbers, 1, 3, 5, &c., and of ascent as these numbers inverted ; 
also, that with the acquired velocity continued uniformly, a body 
moves twice as far as it must descend to acquire that velocity. If 
a body be projected up an inclined plane, it will ascend as far as it 
must descend in order to acquire the velocity of projection. The 
distance passed over in the time t by a body projected with the 

velocity v, down or up an inclined plane, equals t v *-nj 
These statements are proved as in the case of free fall, Chapter II. 

152. Formulae for the whole Length of a Plane. 

1. The velocity acquired in descending a plane is the same as 
that acquired in fatting down its height. j 

For now s = ?; hence (formula 4), v f-^ j (2 g h) , 


which is the formula for free fall through //, the height of the 

On different planes, therefore, v x Ji*\ 

2. The time of descending a plane is to the time of falling down 
its height as the length to the height. 

i 2 

For (formula 2) t = ( ^J = I \j\ But the time of 

fall down the height is ( j . Therefore, 

t down plane : t down height : : I I 7 ) : ( ) 5 


On different planes, t x -. 

It follows that if several planes have the same height, the veloc 
ities acquired in descending them are equal, and the times of 
descent are as the lengths of the planes. For, let A C, A D, A E, 

(Fig. 114) have the same height A J5; then, since v x h~, and his 
the same for all, v is the same. And since t x - , and h is the 
same for all the planes, t x I. 

FIG. 114. 

153. Descent on the Chords of a Circle. In descending 
the chords of a circle which terminate at the ends of the vertical 
diameter, the acquired velocities are as the lengths, and the times of 
descent are equal to each other and to the time of falling through the 

For (Art. 152) the velocity acquired on A C (Fig. 115) = 


(3 a . A c)~ = ( 2 q . r-^r I AC \rw\ ,which, since 

V /i / / \ /I /i / 

is constant, varies as A C, the length. 

(2 ^4 (7 2 V^ 
1 ) = 

fy A B A c\ ^ /2 A B\ ^ 

/ 1 = ( 1 , which is equal to the time of falling 

\ y * y 

freely through A B, the diameter. 

154. Velocity Acquired on a Series of Planes. If no 

velocity be lost in passing from one plane to another, the velocity 
acquired in descending a series of planes is equal to that acquired 
in falling through their perpendicu- Fj& llg 

lar height. For, in Fig. 116, the A E 

velocity at B is the same, whether // ~~s* 

the body comes down A B or E B, *%// ...- 

as they are of the same height, Fb. 
If, therefore, the body enters on B C 
with the acquired velocity, then it is 
immaterial whether the descent is 
on A B and B C or on E (7; in D 

either case, the velocity at C is equal to that acquired in falling Fc. 
In like manner, if the body can change from B C to CD without 
loss of velocity, then the velocity at D is the same, whether ac 
quired on A B, B (7, and CD, or on F D, which is the same as 
down F G. 

155. The Loss in Passing from one Plane to Another. 

The condition named in the foregoing article is not fulfilled. A 
body does lose velocity in passing from one plane to another. And 
the loss is to the whole previous velocity as the versed sine of the 
angle between the planes to radius. 

Let B F (Fig. 117) represent the velocity which the body has 
at B. Eesolve it into B D on the second plane, and D ^perpen 
dicular to it. B D is the initial velocity on B C; FIG 
and, if B I = B F, D I is the loss. But D I is 
the versed sine of the angle F B D, to the radius 
B F; and .*. the loss is to the velocity at B as D I 
: B F : : ver. sin B : rad. 

156. No Loss on a Curve. Suppose now 
the number of planes in a system to be infinite ; 
then it becomes a curve (Fig. 118). As the angle 
between two successive elements of the curve is in 
finitely small, its chord is also infinitely small; but 



its versed sine is infinitely smaller stilly i. e., an infinitesimal of the 

second order; for diam. : chord : : chord : ver. 

sin. Therefore, although the sum of all the 

infinitely small angles is a finite angle, A GD, 

yet, as the loss of velocity at each point is an 

infinitesimal of the second order, the entire 

loss (which is the sum of the losses at all 

the points of the curve) is an infinitesimal of 

the first order. 

Hence, a body loses no velocity on a curve, and therefore 
acquires at the bottom the same velocity as in falling freely 
through its height. 

It appears, therefore, that whether a body descends vertically, 
or on an inclined plane, or on a curve of any kind, the acquired 
velocity is the same, if the height is the same. 

1.57. Times of Descending Similar Systems of Planes 
and Similar Curves. If planes are equally inclined to the hori 
zon, the times of describing them are as the square roots of their 
lengths. For, if the height and base of each plane be drawn, simi 
lar triangles are formed, and h : Zis a constant ratio for the several 

planes. By Art. 152, t <x - oc - : cc VT; that is, the time va- 

Vh Vl 

ries as the square root of the length. 

If two systems of planes are similar, i. e., if the corresponding 
parts are proportional and equally inclined to the horizon, it is 
still true that the times of descending them are as the square roots 
of their lengths. 

(Fig. 119) be similar, and let A .Pand 

af be drawn horizontally, and the lower planes produced to meet 
them, then it is readily proved that all the homologous lines of 
the figures are proportional, and their square roots also propor 
tional. Then (reading t, A B, time down A B, &c.), 

we have t,AB\ t,al) :: VAJB : \Tal)\ 

t,EB:t,el):\ VHTB : V~e~lj : : VTTl : Vab; 


and t,EC\t,ec:: VHTC : Ve~c : : V1TB : 

/. (by subtraction) t,BC\t,lc\\ */~AB : VaU. 

In like manner, t, C D : t, c d : : VA B\ Vab. 
. . (by addition) 

t, (A B + B G + CD) : t, (a I + I c + c d) : : \/~AB : Val 

: : \/(A B + B G + CD) : +/(a I + I c + c d. 

Though there is a loss of velocity in passing from one plane to 
another, the proposition is still true; because, the angles being 
equal, the losses are proportional to the acquired velocities ; and 
therefore the initial velocities on the next planes are still in the same 
ratio as before the losses ; hence the ratio of times is not changed. 

The reasoning is applicable when the number of planes in each 
system is infinitely increased, so that they become curves, similar, 
and similarly inclined to the horizon. Suppose these curves to be 
circular arcs ; then, as they are similar, they are proportional to 
their radii. Hence, the times of descending similar circular arcs 
are as the square roots of the radii of those arcs. 

158. Questions on the Motions of Bodies on Inclined 

1. How long will it take a body to descend 100 feet on a plane 
whose length is 150 feet, and whose height is 60 feet ? 

Ans. 3.9 sec. 

2. There is an inclined railroad track, 2J- miles long, whose 
inclination is 1 in 35. What velocity will a car acquire, in run 
ning the whole length of the road by its own weight ? 

Ans. 106.2 miles per hour. 

3. A body weighing 5 Ibs. descends vertically, and draws a 
weight of 6 Ibs. up a plane whose inclination is 45. How far will 
the first body descend in 10 seconds ? Ans. 3.44 feet. 

4. A body descends vertically and draws another body of half 
the weight up an inclined plane. When the bodies had described 
a space c the cord broke, and the smaller body continued its mo 
tion through an additional space c before it began to descend. 
What is the inclination of the plane ? Ans. 30. 

159. The Pendulum. A pendulum is a weight attached by 
an inflexible rod to a horizontal axis of suspension, so as to be free 
to vibrate by the force of gravity. If it is drawn aside from its 
position of rest, it descends, and by the momentum acquired, rises 
on the opposite side to the same height, when gravity again causes 
its descent as before. If unobstructed, its vibrations would never 

A single vibration is the motion from the highest point on one 
side to the highest point on the other side. The motion from the 

LENGTH OF A PN r, U-L vyt. .< , ^ : 101 

\ < .- :- . .>. / /- / > / *, ;,x 

highest point on one side to the same point again is called a 
double vibration. 

The axis of the pendulum is a line drawn through its centre of 
gravity perpendicular to the horizontal axis about which the pen 
dulum vibrates. 

The centre of oscillation of a pendulum is that point of its axis 
at which, if the entire mass were collected, its time of vibration 
would be unchanged. 

The length of a pendulum is that part of its axis which is 
included between the axis of suspension and the centre of oscil 

All the particles of a pendulum may be conceived to be col 
lected in points lying in the axis. Those which are above the cen 
tre of oscillation tend to vibrate quicker (Art. 157), and therefore 
accelerate it; those which are below tend to vibrate slower, and 
therefore retard it. But, according to the definition of the centre 
of oscillation, these accelerations and retardations exactly balance 
each other at that point. 

160. Calculation of the Length of a Pendulum. Let 

C q (Fig. 120) be the axis of a pendulum in which all its weight 
is collected, G the point of suspension, G the centre of 
gravity, the centre of oscillation, , b, &c., particles FlG - 12( 
above 0, which accelerate it, p, q, &c., particles below 0, T- c 
which retard it. C ^= I, is the length of the pendulum 
required. Denote the masses concentrated in a, b . . . 
p, q, by m, m . . . . m" m ", and their distances from C by 
r, r . . . . r", r "\ and denote the distance from G to G 
by k. Denote the angular velocity by ; then the ve 
locity of m will be r 6 and its momentum will be m r 6. 

If m had been placed at 0, the moving force would 
have been mid. The difference m (I r) 0, is that por 
tion of the force which accelerates the motion of the 

The moment of this force with respect to C is 
f (J - r) r 0. 

In like manner the moment of m is m (I r ) r 0, 
and so on for all the particles between C and 0. 

The moments of the forces tending to retard the sys 
tem applied at the points p, q, &c., are 

m" (r" - I) r" 0, m " (r " - 1) r " 0, &c. 
But since these forces are to balance each other, we have 

m (l _ r ) r Q + m (l - r ) r + &c. = m" (r" - I) r" 
+ m (r 1 " - I) r " 6 + &c.; 

- -o 



mr 2 + m r r - + m"r m + &c. 
whence I - .. - = r -.- i 

m r 4- w r + iwr r" + &c. 

Or I -~ - ,-, where 8 denotes the sum of all the terms similar 
8 (m r) 

to that which follows it. 

The numerator of this expression is called the moment of inertia 
of the body with respect to the axis of suspension, and the denomi 
nator is called the moment of the mass, with respect to the axis of 

By the principle of moments (Art. 77) m r + m r -f &c., or 
S (m r) =M k, where M denotes the entire mass of the pendulum ; 

S (m r 2 ) 
nence, by substitution, I = */-?. 

That is, the distance from the axis of suspension to the centre 
of oscillation is found by dividing the moment of inertia, with 
respect to that axis, by the moment of the mass with respect to the 
same axis. 

161. The Point of Suspension and the Centre of Os 
cillation Interchangeable. Let the pendulum now be sus 
pended from an axis passing through 0, and denote by I the 
distance from to the new centre of oscillation. The distances of 
a,b....p,q, from 0, will be I r, I r , &c., and the distance 
G will be I - k. 

Hence, from the principle just established, we haye 

, _ S[m(l r) 2 ] __ S (m V - 2 m r I 4- m r a ) 
k)~ M(l-k) 

-2S(mrl) + S(mr*) 

M (I - Jc) 

But 8 (m r 2 ) = M Jc I ; and since I is constant, 

__ M?-%lS(mr) +MJcl __ MF - 2 M k I + M Jc I 
M(l-k) M(l-k) 

k} I _ 
k) ~~ 

This last equation shows that the centre of oscillation and the 
point of suspension are interchangeable ; that is, if the pendulum 
were suspended from 0, it would vibrate in the same time as when 
suspended from G. 

162. The Cycloid. One of the methods of investigating the 
theory of the pendulum is by means of the properties of the cy 
cloid. This curve is described by a point situated on the circum 
ference of a circle, as it rolls on a straight line. 

Let the circle^! H B (Fig. 121) make one revolution upon the 



line C A JT, equal to its circumference ; the curve line C D B X, 
traced out by that point of the circle which was in contact with 
C when the circle began to 
roll, is called a cycloid. If 
C X be bisected in A, and 
A B be drawn at right angles 
to it, it is evident, from the 
manner in which the curve 
is generated, that it will have 
similar branches on both 
sides of A B, and that its vertex B will be so placed as to make 
the axis A B equal to the diameter of the generating circle. The 
properties of the cycloid, as applied to the vibration of the pendu 
lum, are the following. 

163. The Cycloidal Ordinate D H equals the circular arc 
B H. For, let Z> D a (Fig. 121) be the position of the circle when the 
generating point is at D ; draw the diameter b a parallel to B A, 
and from D draw DHL parallel to C A ; then the arc D a = arc 
HA, :. the sines D 0, H L, are equal; hence D H L\ but 
from the mode in which the cycloid is generated, C a = arc D a, 
and C A = semi-circumference B HA ; hence D H L a A 

C A C a semi-circumference B HA arc HA = arc B H. 

164. A Tangent to the Cycloid at any point, E (Fig. 122), 
is parallel to the corresponding chord B K of the generating circle. 
Draw DHL infinitely near to E K M ; 

join B K, and produce it to k. The 
elementary triangle H K k is similar to 
the triangle K R B formed by the tan 
gents (K R, B R) to the circle at the 
points .5", B, and is consequently isos 
celes ; .:KH=H k. Now (Art. 163), 
arc B KH = D H\ from which equation 
subtract the previous one, and arc B K 
= Dk. But SLTcBK= EK\ :. EK 

Die. Hence, since E K and D Tc are equal and parallel, E D and 
K k must also be equal and parallel ; and as the tangent at the 
point E may be considered as coinciding with E D, it must 
therefore be parallel to the chord B K. 

Hence the ends of the cycloid meet the base at right angles; 
for the tangent at C is parallel to B A, the axis. 

165. The Cycloidal Arc B E is equal to twice the correspond 
ing chord B K of the generating circle. Draw H o perpendicular 
to K k ; and since the triangle K H k is isosceles, H o bisects the 

FIG. 122. 



base 1\ &, /. K Ic or E D = 2 K o ; and since H o may be consid 
ered as a small circular arc described with radius B IT, K o = Bo 
R K B H B K\ hence E D and K o are corresponding 
increments of the cycloidal arc B E and the chord B K\ and as 
the arc and chord begin together from the point B, and every in 
crement of the former is twice the corresponding increment of the 
latter, the arc B E must be equal to twice the chord B K\ conse 
quently, the whole arc B C twice the diameter A B ; and the 
length of the whole curve C B X (Fig. 121) = 4 A B. And as 
CA X TT . A B 9 therefore the whole cycloid : its base : : 4 : TT. 

166. Descent by Gravity on a Cycloid E F M (Fig. 
123) is a circle whose di 
ameter E Mis perpendic 
ular to the horizon, and 
B G M is the corre 
sponding semicycloid. 

Let the body begin to 
descend from any point A. 

Draw A D parallel to 
B E., and upon M D as a 
diameter describe the cir- * 
cle D N P M 9 with its 
centre at 0. 

Put h = DM,r= CEy z = DH 

FlG - 

the elementary arc G K will be 







then the time of describing 
G K 



dividing (I) by (2), 


/ox AT 
. (2) Now KL PQi hence, 




and the time of descending G K is 

4/9 v> T AT" P 

v r A I Jj J.V JL 

; whence 

In like manner, the time of describing any other elementary 

2 fr 
arc will be found to be T y - times the corresponding arc on the 


circu inference Z>JVJP N; hence the time of describing the cycloid- 

al arc A J/ will be \ f/- x arc D NP M = ~ \/~ . ^ = 
ii g li g A 

This expression for the time down A M being independent of 
//, is very remarkable, for it proves that 

The time of descent on a cycloid to the lowest point is always the 
same, from whatever point in the curve the body begins to descend. 

The.time of falling through E M is 2 y -; .-. time down A M 


: time down E MUTT \ - : 2 \ - :: TT : 2 : : semi-circumference : 

y y 

167. The Involute of a Semicycloid. The involute of 
any curve is another curve described by the extremity of a tangent 
as it unwinds from the former, which 
is called the evolute. If, for example, 
a tangent of a circle unwinds from it, , 
the circumference of the circle is the 
evolute, and the spiral described by the 
end of the tangent is the involute of 
the circle. The involutes of most 
curves are different from their evo- 
lutes; but in the case of the semi- 
cycloid, the involute and evolute are 
of the same form and size. 

Take any line S C (Fig. 124), and 
draw S A at right angles to it ; make 
S C : S A :: semi-circumference of a 
circle : its diameter ; and complete the 
parallelogram S C D A. Produce S A to B, making A B = 8 A ; 
upon S C, A D, describe two semicycloids 8 D, D B, the vertex of 
the former of which is at D, and the latter at B ; then if the tan 
gent unwinds, beginning at D, until the point of contact reaches 
S 9 its extremity will always be found in the semicycloid D B. For, 
through any point F on A D, draw E F G perpendicular to S C, 
and through E draw B G parallel to 8 C\ then E G = S B ; on 
E F, F G, describe the semicircles E T F, F P G, and draw the 
chords T F, F P, the former of which (Art. 164) is a tangent to 
the cycloid 8 D at T. Now 8 E = arc E T, and 8 C = E T F , 
:. OE (= D F) = arc TF; but D F = F P; .-. arcs F T, F P 
are equal, and also the angles subtended, T E F, F G P. There- 




fore, as Tand P are right angles, E F T = P F G, and P F Tis 
a straight line; moreover, TP=2TF= (Art. 165) the cycloidal 
arc T D. Therefore, T P is a tangent unwound from D, and P 
is its extremity ; and P having been assumed anywhere on the 
semicycloid D B, it follows that D P B is the involute of 8 T D. 

168. The Cycloidal Pendulum. A pendulum may be 
made to vibrate in a cycloid by attaching the weight P (Fig. 125) 
to a flexible cord, whose 

point of suspension is at 
S, where two semicy- 
cloids meet. The cord 
and the semi-cycloid 
should be of the same 
length, and then (Art. 
167) the weight P will, 
at each vibration, de 
scribe arcs of the cy 
cloid D B E 7 as involutes 
rfSDsm&SE. Hence, 
the conclusion arrived at 
in Art. 166 applies to tHis 

motion ; namely, the time down P B from any point P : time down 
A B : : semi-circumference : diameter; .-. doubling the antecedents, 
the time of a single vibration : time of fatting half the length of the 
pendulum : : n : 1. 

169. Application to the Circular Pendulum. Since the 
proportion at the close of the foregoing article is always true, from 
whatever point the descent commences, therefore 

All the vibrations of a cycloidal pendulum are performed in 
equal times, however large or small the extent of swing. 

This is not true of any other curve. But it is evident that a 
very short arc of a cycloid at the lowest point B is coincident with 
the arc of a circle whose centre is S. Hence, if a pendulum vibrate 
through very short arcs, the conclusions are practically true, that 
in circular pendulums also unequal arcs are described in equal 
times, and that the time of a vibration is to the time of falling 
through half the length of the pendulum as TT is to 1. For this 
reason, the pendulum of an astronomical clock is so connected 
with the machinery by its scapement, as to vibrate in small arcs. 

170. Relation of Time, Length, and Force of G-ravity. 

Let I = the length of a pendulum, that is, the distance from the 
point of suspension to the centre of oscillation. Then the time of 


falling half its length = ( ) = M . Hence, putting t = time 
of a single vibration, 

Therefore, the length of a pendulum being known, the time of 
one vibration is found ; and on the other hand, if the time of a 
vibration is known, the length of the pendulum is obtained from 

From the same formulae, we find that t oc VT, or 

The time in which a pendulum makes a vibration varies as the 
square root of the length. 

As t oc Vl, .: I oc t* ; hence, if the length of a seconds pendulum 
equals I, then a pendulum which vibrates once in two seconds 
equals 4 I, and one which beats lialf seconds -} I, &c. 

Again, by observing the length of a pendulum which vibrates 
in a given time, the force of gravity, g, may be found. For, as I 

~JTJ 9 -* And ^ 9 varies, as it does in different latitudes 

and at different altitudes, then I = *-? x g f ; and if the time is 

constant (as, for example, one second), then I oc g. Hence, 

The length of a pendulum for beating seconds varies as the force 
of gravity. 

Also, t oc ( 1 ; that is, the time of a vibration varies directly 

as the square root of the length, and inversely as the square root 
of the force of gravity. 

Since the number, n, of vibrations in a given time varies in- 



versely as the time of one vibration, therefore n oc (^-J , and 

g oc I n a . Hence, if the time and the length of a pendulum are 

The force of gravity varies as the square of the number of vibra 

1. What is the length of a pendulum to beat seconds, at the 
place where a body falls 16^, ft. in the first second? 

Ans. 39.11 inches, nearly. 

2. If 39.11 inches is taken as the length of the seconds pendu 
lum, how long must a pendulum be to beat 10 times in a minute ? 

Ans. 117J feet. 

3. In London, the length of a seconds pendulum is 39.1386 



inches ; what velocity is acquired by a body falling one second in 
that place ? Ans. 32.19 feet. 

171. The Compensation Pendulum. This name is given 
to a pendulum which is so constructed that its length does not 
vary by changes of temperature. As all substances expand by 
heat, and contract by cold, therefore a pendulum will vibrate more 
slowly in warm than in cold weather. This difficulty is overcome 
in several ways, but always by employing two substances whose 
rates of expansion and contraction are un 
equal. One of the most common is the grid 
iron pendulum, represented in Fig. 126. It 
consists of alternate rods of steel and brass, 
connected by cross-pieces at top and bottom. 
The rate of longitudinal expansion and con 
traction of brass to that of steel is about as 
100 to 61 ; so that two lengths of brass will 
increase and diminish more than three equal 
lengths of steel. Therefore, while there are 
three expansions of steel downward, two up 
ward expansions of brass can be made to neu 
tralize them. In the figure the dark rods rep 
resent steel, the white ones brass. Suppose 
the temperature to rise, the two outer steel 
rods (acting as one) let down the cross-bar d\ 
the two brass rods standing on d raise the bar 
I ; the steel rods suspended from b let down 
the bar e, on which the inner brass rods stand, 
and raise the short bar c\ and finally, the 
centre steel rod, passing freely through d and e, lets down the disk 
of the pendulum. These lengths (counting each pair as a single 
rod) are adjusted so as to be in the ratio of 100 for the steel to 61 
for the brass ; in which case the upward expansions just equal 
those which are downward, and therefore the centre of oscillation 
remains at the same distance from the point of suspension. 

If the temperature falls, the two contractions of brass are equal 
to the three of steel, so that the pendulum is not shortened by 

The mercurial pendulum consists of a steel rod terminating at 
the bottom with a rectangular frame in which is a tall narrow jar 
containing mercury, which is the weight of the pendulum. It 
requires only 6.31 inches of mercury to neutralize the expansions 
and contractions of 42 inches of steel. 



FIG. 127. 



172. Path of a Projectile. It has been shown already 
(Art. 44), that a body projected in any direction not coincident 
with the vertical, describes a parabola. In swift motions, however, 
the path of a projectile differs widely from a parabola ; and the 
laws of atmospheric resistance must be employed to obtain cor 
rections for the conclusions deduced in this chapter. 

173. Formulae Investigated. In order to investigate the 
general formula, let A (Fig. 127) be the point of projection, A B 
the plane over which the body is projected, passing through A. 
A B also denotes the range or dis 
tance to which the body is thrown. 

Let A C be drawn parallel, and 
BCD perpendicular to the hori 
zon. Put a G A D, the angle of 
elevation ; b = C A B, the angle of 
elevation or depression of the plane 
of the range ; v = the velocity of 
projection ; t = the time of flight ; 
r the range ; and g = 32 J feet, 
the velocity imparted by gravity in 
one second. 

Then, by the laws of uniform motion, at the end of the time t, 
if gravity did not act, the body would be found in the point D, 
while, by the laws of falling bodies, it would in the same time pass 
through the perpendicular D B ; consequently, 

A D = tv ; SLn&DB=-\gt\ 

In the right-angled triangles ABC and ADC, the angle B 
is the complement of b, and the angle D is the complement of a ; 
and, since the sides are as the sines of the opposite angles, 

cos I : sin (a V) : . tv : tv - 

Plus is used when the plane A B descends ; minus, when it 

gt _ sin (a b) 
2 v ~ cos b 



. . . , r sin (a ; I) 

Again, cos a : sin (a b) : : r : - - \ g t\ 

cos a 



. . 
2 r cos a 

Eliminating t from (1) and (2), we have 

r _ 2 sin (a =t b) cos 
v 2 # cos b 

From these three equations, all the relations between the time, 
Telocity, range, and angle of elevation, are readily determined; 
so that any two of these four quantities being given, the other two 
may be found. Thus, 

D 4.- /-i\ gtcosb 

By equation (1) v = ? -, -- rr. 
2 sin (a b) 

T> j.- /n\ ( ) P COS # 

By equation (2) r ^-4 1 ^-^ 
2 sin (a =fc J) 

The wmgre and elevation being given, to find the fo me and 

/2 r sin (a db J)\ 2 

By equation (2) = ( - x - \ . 

\ g cos a) I 

/o\ / r 9 cos2 ^ \ 2 

By equation (3) = ^ ^ ^ &) ^ J . 

The velocity and elevation being given, to find the fo me and 

, . , N 2 v sin 

By equation (1) t = - 

^ cos 

, . /ox 2 v 2 sin (# J) cos 

By equation (3) r - - rr^- 

^ cos 2 1 

If any two of the above quantities be given to find the angle of 
elevation, then (b being known) in order to find the value of a we 
substitute in formulae (1), (2), and (3), for sin (a #), its value, 

viz., sin a cos b sin b cos a, and, in reducing, put tan for . 


Formula (1) becomes sin a tan b cos a J-^, whence by 

eliminating cos a, the value of sin a can be found. The resulting 
equation being a quadratic, there will be, in general, two values 
of sin a ; that is, two angles of elevation for the same value of 
v and t. 

Formula (2) becomes tan a cos b sin b = -, whence 

tan a = 5- - ^ rp tan 5. 
2 r cos ^ 

/y A* 
Formula (3) becomes sin a cos a tan 5 cos 2 a = y 



Put c = , and x = sin , then (1 z a )~ = cos ; and 

ii V 

x (I x*Y tan 1) (1 a; 2 ) = c, from which x or sin may be 

174. Different Angles of Elevation for the Same 
Range. As this last equation is a biquadratic, it will give four 
yalues of x ; the two positive values indicate that there are two 
different angles of elevation corresponding to the same values of 
v and r. When these two values are equal, then, as shown 
below, a = % (90 =F b), in which case the range (r) is a maxi 
mum; and there is the same range for any two angles equally 
above and below that which gives the maximum. For, since 

r = 2 v* Sm ^ /r , if v and the angle b are given, the range 


will vary as sin (a b) cos a. But 

sin (a b) cos a = sin a cos a cos b sin b cos 2 a 

= sin 2 a cos 5 sin 5 (A + i cos 2 ) 

= A sin 2 n cos # cos 2 a sin & db 4 sin 5 
= sin (2 b) A sin 5 ; 

and since the second part of this expression is constant, the range 
will be a maximum when sin (2 a db b) is a maximum ; that is, 
when 2 a b = 90. 

.-. a = -1 (90 =p 5). 

Therefore the range will be a maximum when the angle of eleva 
tion is equal to 4 (90 =F b). 

a sin (90 + 2 c) sin b 
When a = t (90 ^ b) + c, r = v a - ^ cos . ft" -, 

. sin (90 - 2 c) sin 
and when = i (90 T &)-<?, r = v*- ^ cos & 

But sin (90 + 2 c) = sin (90 2 c), since the sines of sup 
plementary arcs are equal ; hence all angles of elevation, equally 
above and below that which gives the maximum, have equal ranges. 
Thus, a cannon ball fired at an angle of 60 above a horizontal 
plane, would reach the plane at the same distance from the point 
of projection as if fired at an angle of 30. When the data of the 
problem give or require a greater value for sin (2 a b) than 1, 
the sine of 90, the problem, under the proposed conditions, is 

That the two values of sin a are equal when the range is a 
maximum, may be shown as follows : 

Let x and y be two varying supplementary arcs, and let 

a sin x sin b 
a = J ( X zp b) ; then r = - i 


* . -, * 1 , , x ,, 2 sin y d= sin ft 

Again, let = J (y =p ft) ; then r = v* __. 

Now, although these two values of may be different, yet the 
ranges corresponding to them will be equal, because sin x sin y. 

Suppose x to increase from to 90 ; then y will decrease from 
180 to 90, and the two values of a will become equal, each being 
4 (90 =p ft). But, as has been shown, this value of a gives a maxi 
mum for r. 

175. The Greatest Height of a Projectile. To find the 
greatest height to which th projectile will ascend, it must be con 
sidered that a body projected perpendicularly upward, will rise to 
the same height from which it must have fallen to acquire the 
velocity of projection (Art. 24). Since v represents the whole 
velocity of projection in an oblique direction, and since a is the 
angle of elevation, therefore v sin a is that component of the 
velocity which acts directly upward. And the space described 

vertically by tnis component of the velocity is [(3) Art,28] s = . 

Hence, substituting h for s, and v sin a for v, we have 

If, therefore, the angle of elevation and the velocity of projec 
tion are given, the greatest height is found as above. Or, if the 
angle of elevation, and that of the plane, be given, along with the 
range (r) or the time (t), then let v be found first, as in Art. 173 ; 
after which h may be obtained from equation (4). 

If the velocity of projection, and the greatest height to which 
the projectile rises, were given, equation (4) will determine the 

. v* sin 2 a . 2 a h 
angle of elevation. For since h = ^ , .*. sm a = ~- , and 


sin a = . 


176. Particular Formulae for a Horizontal Plane. The 

preceding equations become much more simple when the projec- 
. tion is above a horizontal plane ; for then ft = ; therefore sin ft 
0, and cos ft = 1 ; hence, from equations (1), (2), and (3), we 

, 2 v sin a . ,- ,\ 

t = - - cc v sm a (1 ), 



On a horizontal plane, therefore, we have the following the 
orems : 

I. The TIME OF FLIGHT varies as the velocity of projection mul 
tiplied Inj the sine of the angle of elevation. 

II. The RANGE varies as the square of the velocity of projection, 
multiplied by the sine of twice the angle of elevation. 

Moreover, since the sine of twice 45 equals the sine of 90, 
which equals radius, hence, hy Theorem II, 

III. The RANGE is GREATEST when the angle of elevation is 
45, and is the same at elevations equally above and below 45. 

IV. The TIME OF FLIGHT is GREATEST when the body is 
thrown perpendicularly upward. 

177. The Equation of the Path of a Projectile. Sup 
pose the body is projected from A (Fig. 128) in the direction A T 
with a velocity v, and let 
A X, horizontal, and A Y, 
vertical, be rectangular 

The components of v 
along the axes are, v cos a 
for A X, and v sin a for 
A Y. At the end of the 
time #, suppose the body to 
be at C. Denote the co-or 
dinates of the point C by x 
and y ; then x = t v cos #, 
and y B D B C = tv sin a ^ 

Eliminating t, we have 

= x tan a 

, 5, 
2 v* cos 2 a 

an equation expressing the relation between x and y for any value 
of t whatever, and consequently the equation of the path. 

To find the range A E, make y = 0; then x 0, or x 
2 v* sin a cos a 

The first value of x corresponds to the point A ; the second is 
the range A E. 

To find the time of flight, make x = r in the equation x = 


t v cos , and we have t - 

v cosa 

(1 T 8 

If a = 0, y = ~-; a , the equation of the path when the body 

is thrown horizontally. Since y is negative for all values of x, 
every point of the path, except A, lies below a horizontal line 


drawn through the point of projection. A G f represents the path 
when the body is projected in the line A X. 

If positive ordinates are estimated from A X downward, the 

Cl OT^ 

equation may be written y = |-^. 

178. To Find the Range on an Oblique Plane. Let b 

be the inclination of the plane to the horizon ; then y x tan b 
is the equation of the line in which the oblique plane intersects 
the plane of the projectile s path. Combining this with the equa- 

Cl 3? 

tion if = x tan a t~- , we have x tan b = x tan a 
2 v cos a 

a x* 2 v* cos 2 a ,, 

f ; whence x = 0, and x- - (tan a =F tan b) 

2 v* cos 3 a g 

2 v* cos a sin (a ^ b} -, , ., .,, , x 

, and hence the range will be r = =- 

g cos b cos b 

2 v* cos a sin (a =F b) 
g cos 2 b 

179. Questions on Projectiles. 

1. A gun was fired at an elevation of 50, and the shot struck 
the ground at the distance of 4898 feet ; with what velocity did it 
leave the gun, and how long was it in the air ? 

Ans. Velocity, 400 feet per second. 
Time, 19.05 seconds. 

2. Eange 4898 feet, time of flight 16 seconds; required the 
angle of elevation and the velocity of projection ? 

Ans. a = 40 3 , v = 400 feet per sec. 

3. Eange 2898 feet, velocity of projection 389.1 feet, what were 
the elevation and time of flight? 

Ans. a = 19 or 71, t = 7.87 or 22.86 sec, 

4. Elevation 40, range 4898; required the range when the 
elevation is 29^ Ans. 4263. 

5. Elevation 40 3 , time of flight 16 seconds ; required the 
range and velocity of projection ? Ans. r = 4898, v = 400 ft. 

6. Velocity 510 feet per sec., time of flight 15 seconds, to find 
the elevation and range. Ans. a = 28 14 , r = 6740. 

7. On a slope ascending uniformly above a horizontal plane at 
an angle of 10 20 , a ball was fired at an angle of elevation above 
the horizon of 34, and with a velocity of 401 feet per second ; 
what was the range on the slope when the gun was directed up the 
hill, and what when directed downward ? 

Ans. 3438 and 5985 feet 

8. What will be the time of flight for any given range, the 

angle of elevation being 45 ? 

Ans. t 



9. Having given the angle of elevation, to determine the veloc 
ity, so that the projectile may pass through a given point. 

x / # 

Ans. v = V ^-7-r r> where x 1 and y are 

cos. a * 2 (* tan a - y y 

the co-ordinates of the given point. 

10. Find the angle of elevation and velocity of projection of a 
shell, so that it may pass through two points, the co-ordinates of 
the first being x = 1700 ft, y = 10 ft, and of the second, x" = 
1800 ft, y" = 10 ft Ans. a = 39 19", v = 2218.3 ft 


180. Central Forces Described. Motion in a curve is 
always the effect of two forces ; one an impulse, which alone would 
cause uniform motion in a straight line ; the other a continued 
force, which urges the body toward some point out of the original 
line of motion. The first is called the projectile force, the second 
the centripetal force. 

The centripetal force may be resolved into two components; 
one in the direction of the tangent, the other perpendicular to it. 
The tangential component will accelerate or retard the motion in 
the curve according as it acts witli the projectile force, or in oppo 
sition to it. When the body moves in the circumference of a 
circle, the tangential component of the centripetal force is 0, and 
hence the motion is uniform. 

If the centripetal force should cease to act at any instant, the 
body, by its inertia, would immediately begin to move in a straight 
line tangent to the curve at the point where the body was when 
the force ceased to act. 

Since the body, by its inertia, tends to move in a tangent, there 
is a continued outward pressure directed from the centre of curva 
ture ; this is called the centrifugal force. In circular motion it is 
equal to the centripetal force, and directly opposed to it. 

181. Expressions for the Centrifugal Force in Circular 

1. Let r = the radius of the circle, r = 
the velocity of the body, c = the distance 
through which the centrifugal force causes 
the body to move in one second, and let 
A B (Fig. 129) be the arc described in the 
infinitely small time t ; then A B v t, 
and, by a method similar to that employed 
in the discussion of the force of gravity, it 
may be shown that B D = c t\ 


But A B, being a ver} T small arc, may be considered as equal 
to its chord, which is a mean proportional between A E and the 

v 2 f 

diameter 2 r. Hence c t* = -^ , or 


If this be doubled, then (Art. 25) is the velocity which the 

centrifugal force is capable of generating in one second, and this is 
sometimes taken as the measure of the centrifugal force. 

From (1) it follows that in equal circles the centrifugal force 
varies as the square of the velocity. 

2. The value of c may be expressed in a different form. Let 
i = the time of a complete revolution ; then 2 TT r = v t r ; whence 

v = 77. This substituted in (1) gives 

. = *- ... ....... ; 

Hence the centrifugal force varies directly as the radius of the 
circle, and inversely as the square of the time of revolution. 

3. Let w the weight of the revolving body, and c = the 
centrifugal force expressed in pounds ; then 

w.c i-i^gin ; whence c = --- . . (3) 

Let n = the number of revolutions per second ; then 
i) = 2 IT r n, and (3) becomes 

. 4rr a 
c -- . w . r . n ....... (4) 

182. Two Bodies Revolving about their Centre of 
Gravity. Let A and B (Fig. 130) be two bodies connected by 
a rod, and let them be made to 
revolve about the centre of 

gravity (7; then by (4) the 
centrifugal force of A will be 

-.A.AC. n\ and of B, . B . B C . n\ 

g 9 

But C being the centre of gravity of the two bodies, A . A C = 
B.. B (7; /. the centrifugal force of A equals that of B. Hence 
if two bodies revolve in the same time about an axis passing through 
their centre of gravity, there will be no strain upon that axis. 

183. Centrifugal Force on the Earth s Surface. As the 

earth revolves upon its axis, all free particles upon it are influenced 
by the centrifugal force. Let N 8 (Fig. 131) be the axis, and A a 



particle describing a circumference with the radius A 0. Put r = 

C Q, r = A 0, I = the angle .1 C ft 

the latitude, c the -centrifugal force 

at the equator, c = the centrifugal 

force at A, v = velocity of Q, and v 

= velocity of A ; then 

v i 

But v : V : : r : r ; whence v 1 

Again, from the triangle AGO 

we have r 

v cos I, and e 

r cos I , hence v 1 = 
:os 2 1 v* cos I 


Comparing this value of c 

2 r cos 
with that of c, we have 

c = c cos I. 

That is, 7*e centrifugal force at any point on the earth s surface is 
equal to the centrifugal force at the equator, multiplied ~by the cosine 
of the latitude of the place. 

Let A B represent the centrifugal force at A, and resolve it 
into A D on C A produced, and A F, tangent to the meridian 
N Q S , then, since the angle D A B = A C Q = I, we have 

A D = A B cos I c cos I . cos I = c cos 2 /. 

That is, that component of the centrifugal force at any point which 
opposes the force of gravity is equal to the centrifugal force at the 
equator, multiplied by the square of the cosine of the latitude of 
the place. 

In like manner we find A F A B sin I c cos I sin I = 

. From this equation we see that the tangential com- 

ponent is at the equator, increases till I 45, where it is a 
maximum ; then goes on diminishing till I = 90, when it again 
becomes 0. 

The effect of A D is to diminish the weight of the particle, 
while the effect of A Fis to urge it toward the equator. 

184. Examples on Central Forces. 

1. A ball weighing 10 Ibs. is whirled around in a circumference 
of 10 feet radius, with a velocity of 30 feet per second. What is 
the tension upon the cord which restrains the ball ? 

Ans. 28 Ibs. nearly. 

2. With what velocity must a body revolve in a circumference 
of 5 feet radius, in order that the centrifugal force may equal the 
wight of the body ? Ans. v = 12.7 ft. 



3. A ball weighing 2 Ibs. is whirled round by a sling 3 feet long, 
making 4 revolutions per second. What is its centrifugal force ? 

Am. 117.84 Ibs. 

4. A weight of 5 Ibs. is attached to the end of a cord 3 feet long 
just capable of sustaining a weight of 100 Ibs. How many revo 
lutions per second must the body make in order that the cord 
may be upon the point of breaking ? Am. n = 2.3 nearly. 

5. A railway carriage, weighing 7 tons, moving at the rate of 
30 miles per hour, describes an arc whose radius is 400 yards. 
"What is the outward pressure upon the track ? Am. 786 + Ibs. 

6. A hemisphere has its base fixed in a horizontal position, 
and a body, under the influence of gravity, moves down the con 
vex side of it from the highest point. How far from the base will 
the body be when it leaves the surface of the hemisphere ? 

Am. r. 

185. Composition of two Rotary Motions. 
When a body is rotating on an axis, and a force is applied wliicli 
alone would cause it to rotate on some other axis, the body will com 
mence rotation on an axis lying betiveen them, and the velocities of 
rotation on the three axes are such, that each may be represented by 
the sine of the angle between the other two. 

Suppose that the body H K (Fig. 132) is rotating on A B, 
and that a force is applied to make it 
rotate on G D. Let these axes intersect 
within the body, and call the point of 
intersection, G. Imagine a perpendicu 
lar to the plane of the axes to be drawn 
through G, and let P be a particle of 
the body in this perpendicular. Sup 
pose the particle P, in an infinitely small 
time t, to pass over P a by the first rota 
tion, and P c by the second. Then, 
since the particle will describe the diago 
nal P e in the time t, this line must in 
dicate the direction and velocity of the 

resultant rotation. Therefore, if E F be drawn through G, per 
pendicular to the plane G P e, E F is the axis on which the body 
revolves in consequence of the two rotations given to it. Since 
P G is perpendicular to the plane A G C, and also to the line E F, 
therefore E F is in that plane ; that is, the new axis of rotation is 
in the plane of the other two axes. The angles AGE and E G C, 
are respectively equal to the angles a P e and e P c, the inclina 
tions of the planes of rotation. But the lines, P a, P c s P e, 
represent the velocities in those directions respectively; and 
(Art. 41) P a : P c : P e : : sin c P e : sin a P e : sin a P c\ there- 




fore P a : P c : P c : : sin C G E : sin A O E : sin A G C\ ov, 
the velocities on the three axes, (namely, the axes of the compo 
nent rotations, and of the resultant rotation,) are such, that each 
may be represented by the sine of the angle between the other 
two axes. 

186. The Gyroscope. The gyroscope affords an illustration 
of the composition of two rotations imparted to a body. As 
usually constructed, it consists of a heavy wheel G H (Fig. 133), 
accurately balanced on 
the axis a b, which runs 
with as little friction as 
possible upon pivots in 
a metallic ring. In the 
direction of the axis, 
there is a projection B 
from the ring, having a 
socket sunk into it on 
the under side, so that it 
may rest on the pointed 
standard, S, without 
danger of slipping off. 

The wheel is made 
to rotate swiftly by draw 
ing off a cord wound 
upon a b, and then the 
socket in # is placed on 
the standard, and the whole left to itself. Immediately, instead 
of falling, the ring and wheel commence a slow revolution in a 
horizontal plane around the standard, the point A following the 
circumference A E F, in a direction contrary to the motion of the 
top of the wheel 

This revolution is explained by applying the principle of com 
position of rotations given in the preceding article. The particles 
of the wheel are rotating about the horizontal axis a b by the force 
imparted by the string. The force of gravity tends to make it 
fall, that is, to revolve in a vertical circle around the axis C D at 
right angles to a b. Hence, in a moment after dropping the ring, 
the system will be found revolving on an axis which lies in the 
direction E B, between A B and C D, the other two axes. Now, 
gravity bears it down around a new axis perpendicular to E B. 
Therefore, as before, it changes to still another axis F B, and thus 
continues to go round in a horizontal circle. 

The only way possible for it to rotate on an axis in a new posi 
tion, is to turn its present axis of rotation into that position. 


Hence, the whole instrument turns about, in order that its axis 
may take these successive positions. 

The change of axis is seen also by observing the resultant of 
the motions of the particles at the top and bottom of the wheel. 
For example, G is moving swiftly in the direction in by the rota 
tion around a b ; by gravity it tends to move slowly in the line r, 
tangent to a vertical circle about the centre B. The resultant is 
in the line n, tangent to the wheel when its axis a 1} has taken the 
new position E B. 

The centre of gravity of the ring and wheel tends to remain at 
rest, while the resultant of the two rotations carries around it all 
other parts, standard included, in horizontal circles. But the 
standard by its inertia and friction resists this effort, and the reac 
tion causes the ring and wheel to go around the standard. 



187. Longitudinal Strength. Strength is the power to resist 
fracture ; stress, the power to produce fracture. When a force is 
applied to a bar or rod, to pull it asunder, its strength, called in 
this case longitudinal strength, is proportional to the area of its 
cross-section. Each line of particles in the direction of the length 
of the bar has its separate strength, and the whole strength there 
fore depends on their number, that is, on the area of the cross- 
section. The form of the cross-section is immaterial. 

188. Lateral Strength, Support at Each End. When a 

beam rests horizontally, supported at both ends, and pressed by- a 

weight at the centre, its strength at that point varies as the area 

of the cross-section, multiplied by the depth of the centre of gravity. 

Let A B C D (Fig. 134) represent a longitudinal section of a 

FIG. 134. 





cL D 

i ~I 


prismatic beam and E Tc d I a section of any form whatever, at 
right angles to the axis of the beam. Let G be the centre of 
gravity of the cross-section, and g h, Jc I, m n, &c., the width of 



horizontal lamina? of the beam. Suppose this beam to be sup 
ported at its two ends, and that on the middle of it at E there is 
placed a weight, W. From E draw E d perpendicular to the 
horizon, and cutting the several laminae in the points #, #, c, d, &c. 

The pressure of the weight W tends to produce a fracture in 
the beam, beginning at d, and passing through the larninaB in 
succession until it arrives at E. 

The tendency of the beam to resist fracture depends partly 
upon the cohesion of its corresponding particles, and partly upon 
the distance from E at which the force of cohesion acts. E may 
then be considered as the centre of motion of a lever, at the ex 
tremities of whose arms Ed, E c, E b, E a, &c., this force is applied. 
Let s the cohesive force of one line of particles, then s x y k 
will represent the strength of the lamina whose width is g li y 
and s x g li x E a will represent the power of this lamina to 
resist fracture ; hence the whole power of the beam to resist frac 
ture will be 

- mn x EC + &c.) 

Put A = the area of the cross-section and G = the depth of 
the centre of gravity below E ; then (Art. 78) 

, (jhxEa-{-lclxEb + mnxEc-{- &c. 

(JT = -j 

or A . G g h x E a + kl x E b + mn x EC + &c., and hence 
the strength of the beam is equal to s . A . G <x A . G. 

189. Special Cases. This proposition is general, and ap 
plies to a number of distinct cases. In cylindrical and square 
beams, since the area of the section varies as the square of its 
depth, and the distance of the centre of gravity from the point E 
varies as the depth, their strength is as the cube of the depth. In 
beams whose cross-section is a rect 
angle (Fig. 135), the strength varies as 
the breadth and square of the depth ; 
for here the area being as the product 
of the two sides, and the distance of 
the centre of gravity from E being 
equal to half the vertical side, 
and therefore proportioned to that 
side, the proposition is, that the Jl 
strength varies as the breadth x depth 

x depth, or as the breadth into the square of the depth. Hence, 
the same beam with its narrow side upward, is as much stronger 
than with its broad side upward, as the depth exceeds the breadth. 

FIG. 135. 


For the area being the same in both cases, the strengths are pro 
portioned to E G and Eg, or as A B to A C. Thus if a joist be 
10 inches broad and 2% thick, it will bear four times as much 
weight when laid on its edge as when laid on its side. Hence the 
modern mode of flooring with thin but deep pieces of timber. 

Again, a triangular beam is twice as strong when resting on a 
side, as when resting on an edge. .For, the ai?ea being the same in 
both cases, the strength 

varies as E G and E g FlG - 136. 

(Fig. 136), which are as 
2 to 1 (Art. 72). These 
principles apply not only 
to beams, but to bars, and 
similar forms of every sort 
of matter. 

190. Stress from the Weight of the Beam. The stress 
arising from the weight of a beam varies as the product of the 
length and weight. Let L length, and W weight of the whole 
beam, w = weight of the 
portion A C (Fig. 137). FlG - 137- 

The pressure on each 
prop is {> W , this, there 
fore is the force acting at 
A which tends to fracture the beam at C with an energy expressed 
by $ W x A G. 

Now suppose the portion B to be held firmly in solid ma 
sonry, and a force 4 W to act upward at A, and another = w at 
the middle of A C to act downward, the tendency to produce frac 
ture will be the same as before, and hence the stress at C will be 
$WxA C-wx$AC=t(W-w)AC. But A B : A C:: W: w\ 

A C 

whence w = -^^ Wj so that the expression for the stress will be 
A Jj 

B- A 


A B being constant, the stress at any point C varies as the rectan 
gle of the two lines A C and B C, and is greatest when A C = B C, 
or when C is at the middle of the beam, where the stress is 

If, therefore, we use the term relative strength to denote the ratio 
of the strength to the stress, and represent this ratio by S, we 
shall have 



If beams are similar, then the above ratio is inversely as any 
one of their three dimensions. Being similar, their length, breadth, 
and thickness are proportional. Let D represent any one of these 
three dimensions. Then, 

Since A oc Z> a and G oc D, 
A . G oc Z) 3 ; also, L oc D, and W oc 3 ; 
Z) 3 1 

Hence the relative strength of large structures is less than that 
of smaller similar ones. If a model is three feet long, and the 
structure 75 feet long, then the structure is 25 times weaker rela 
tively than the model. 

191. Additional Weight at the Centre. If a weight W 
is uniformly distributed through a beam whose length is L, the 
stress arising from the weight is (Art. 190) I L . W. If the same 
weight is placed at the middle of the beam, the stress is \ L . - W 
\L IF ; hence, if a weight is placed at the middle point, the 
stress is twice as great as when distributed uniformly through the 

If IF is the weight of the beam, and a weight IF is placed at 
the middle, then, from what has just been shown, the relative 
strength will be 

A.O A. G 

" } L (J W + IF ) L (i W + IF ) 

If the beam is small compared with the weight laid upon it, 

A. G 

In order that the foregoing general formula may be applied to 
practice, so as to find the actual strength of bars or beams, it is 
necessary to have some standard of strength ascertained by experi 
ment, which may be employed as the unit of comparison. For 
example, it is found by experiment that a stick of oak, one foot 
long and one inch square, is able, when supported at both ends, to 
sustain a weight of 600 pounds ; and that a bar of iron of the same 
dimensions would sustain, in the same circumstances, 2190 pounds. 
The oak weighs half a pound, and the iron three pounds. With 
these data applied to the foregoing formulae, we may solve such 
problems as the following: 

1. What weight can be sustained at the middle point of a pris 
matic beam of oak, whose length is 6 feet, and its end 4 inches 
square ? 


If the weight of the bar is left out of account, 

A.G I 2 . A , 4 2 . 2 , , 

one case an( ^ f r ^ ne other. 

L.W ~ 1 . 600 x * 6 . JF" 

But these expressions are to be equal at the moment of rupture of 
both beams, since at that moment they have the same relative 
strength ; 

If the weight of the beams be considered, then 

! - MiTeoo) = 6(aHTr) ; w> = 6378i lbs- 

In this example, the weight of the large beam is known to be 
48 Ibs., from the given dimensions and weight of the small one. 

2. What must be the depth of a beam in the form of a rectan 
gular prism, whose breadth is 2 inches and length 8 feet, to sup 
port a weight of 6400 pounds, its own weight not being taken into 
consideration ? 

3. What weight can be supported at the middle point of a bar 
of iron 10 feet long, and the side of whose square end is 3 inches, 
its own weight not being taken into consideration ? 

Ans. 5913 pounds. 

192. Additional "Weight, at Any Point. The stress pro 
duced by a weight, at any point, is as the product of the two dis 
tances from the ends. In Fig. 138, ac 
cording to the theorems for parallel FIG. 138. 

W x B C A c 

forces, pressure at A = 

W x A C 
pressure at B 7-5 

action of either point of support is equal ^ w 

to the pressure on that point; and this force acts at C with a 

leverage A C on one side, and B C on the other, so that the stress 

WxBC Wx AC n ... ,. , 

at C = -,--= x A.G, or -p- = x B. C, either of which 

-a. JD _/l JJ 

expressions stress at C, and oc A C x B C. And since this 
rectangle is greatest when A C = OB, and diminishes as these 
lines become more and more unequal in length, so the tendency 
of a horizontal bar to break is greatest in the middle, and decreases 
toward tho points of support. 

193. Form f :r Equal Strength. Hence a beam, in order 
to be equally strong throughout, must be thickest in the middle ; 



and if the sides of such a beam are parallel planes, the figure of the 
beam must be elliptical. 


For let the curve A P D M (Fig. 139), whose axis is A 
resent a longitudinal section, 
and let a the thickness or 
breadth of the beam ; then a 
section of the beam perpen 
dicular to the axis at any point 
C will be a rectangle, whose 
breadth is , depth P M, and 
the depth of its centre of gravity \ P M. Hence, the tendency of 
the beam to resist fracture at any point C is as a x P M 2 ; but 
the stress at C is as A C x C D ; therefore 

a x P M* P M 

the relative.strength at C * -^-^ cc AGxCD ; 

hence, if P J/ 3 oc A C x CD, the strength will be the same at 
every point : but in this case the curve A P D M is an ellipse, 
whose major and minor axes are A D and F K. 

Therefore, in the use of horizontal rectangular timbers in 
building, much of the timber is useless, though it would only be a 
waste of labor to remove the redundant parts. But iron beams 
are often made of curved forms, which combine strength with 
lightness and economy of material. The convex form has some 
times been adopted in the iron bars for railroad tracks, as shown 
in Fig. 140. 

FIG. 140. 

FIG. 141. 

194. Lateral Strength, Support at One End. When a 
prismatic beam is secured firmly in 
a wall at one end, the same general 
statements hold true as in relation 
to beams supported at both ends. 

1. The strength varies as the 
area of tlie cross-section multiplied 
l)ij the height of the centre of gravity. 

~LetABEF,at>ef (Fig. 141), 
represent the longitudinal sections 
of two prismatic beams fixed hori 
zontally into the wall H K L M\ 
then the tendency of these beams to 
resist- fracture at the ends E F, e /, 



where they are inserted into the wall, will be measured "by the area 
of the cross-section into the height of its centre of gravity ; for in 
this case the fracture will begin at the upper points F, /, and end 
at the lower points E, e ; that is, strength x A . G. 

But the tendency to produce fracture will be the weight of the 
beams, acting at the distance of their centres of gravity from the 
ends E F, ef. Hence, 

A. a 

2. If an additional weight lies on the end of the beam, then 

A. G 

L (4 W+ WJ 

3. If beams are similar, S x 

D m 

If any other cross-section be taken, as C D, and W represent 
the weight from it to the end A B, then, by the same course of 

A . G 
reasoning as before, the relative strength at G D (= 8) x . 

A. L> . \\ 

If W represent a weight at the end, the strength to support 
W 9 at any section C D, is as A ,/ TJ/ , ; and if W is constant, 

A C. W 

A. G 

A G 

FIG. 142. 


195. Forms of Equal Strength at Every Section. 

1. A beam supported at one end, and having the form of a 
wedge, whose triangular sides are 

parallel to the horizon, has equal 
strength at every section, for sup 
porting a weight at the extremity. 
Let the wedge, in Fig. 142, have the . , 
uniform depth d\ then G = ^ d , 
and the area of the section at C is 
E D . d ; .*. A . G = i E D . d* x 
E D x A G\ hence the relative 

A C 
strength is as -TTT? that * s > ^ * s 



2. A beam, whose vertical sides 
are parallel, and whose longitudinal 
section, parallel to the sides, is a 
semi-parabola, has equal strength at 
every section. In Fig. 142, let d 

the uniform thickness; then A = d . Z> , G = J C D \ .: A . G 


C D * 

= 3 ,d . C D 9 <x C D n . Hence, 8 ~f~ ; but A O x C l> n : 


which is constant 

196. Prismatic Beams Breaking by their own Weight. 

Suppose beams of prismatic or cylindrical form to have similar 
cross-sec fcions, one dimension of which is Z>, and to be supported 

at both ends, or only at one end ; then 8 x 77JT~TT~ }yt\ x 

D 3 

- - . This variation, thrown into the form of a full 
L (lr H 4- H ) 

proportion, becomes 8 : s : : L ^ w+w>) ^T^jTSTy For ex 

ample, let the beams be cylinders, whose lengths are L, I ; their 
diameters, Z>, d\ weights, W, w; while W, w , are additional 
weights laid on their middle points, or the unsupported ends. 
Then the above proportion gives their relative strength. Now 
let the second beam have no weight laid upon it ; that is, let 

-Tfrjc TI But since 10 : W : : I : L, . . w = ^ ; 

1 2 L 

hence S : s : : - (1 , : -^-y a ; which expresses their relative 

strength, when the diameters are equal, and the second beam 
is not loaded. If their lengths and weights become such as 

to cause the beams to break, then, since S = s, . . -7^77-^ jv /7 r 

= _..,t> = _ ___ and? = -_--; which 

gives the length of a beam that breaks by its own weight. 

Let the prismatic or cylindrical beams be simitar to each 

other; then D* : d* :: L* : F; . . S : s : : T -, : ~. But 

A W + Vv 4>w 

the weights of similar solids of the same density are as the cubes 

W . P 

of their homologous dimensions ; /. w : W : : V : L 3 ; /. w = ^ ; 


L* 2 L 3 

hence, by substitution, S : s : : r^ rp> : -^ f And when the 

j tf T" fr W t 

L? 2 L 3 

beams break, since S = 8, therefore -p== , = ^7 7 ? i ^+ ^ 

2 f \~ Vr I 

W.I . L(W+2W) - M 

= -~-T-, or I = - ^ == -. If, therefore, a cylindrical beam 

& LJ n 


whose length is L breaks with the given weight IF placed upon 

it, a similar cylindrical beam whose length is - -=. will 

break with its own weight. 

The same reasoning is applicable to a beam of any prismatic 
form, whatever the shape of the cross-section. 

197. Comparison of Beams Supported at One End 
and at Both Ends. If a horizontal beam be supported at loth 
ends, the stress produced by its own weight, IF, is measured by 
| L x W (Art. 190). 

If the beam be supported at one end only, the stress is measured 
by the whole weight applied at the centre of gravity, and conse 
quently the stress = L x W. 

Therefore a beam supported at both ends has four times the 
relative strength of the same beam supported only at one end. 
And if a certain beam supported at one end breaks by its own 
weight, a beam of the same dimensions twice as long will break 
by its own weight when resting on two supports. 

If, however, instead of the weight of the beam itself, this is left 
out of the account, and a weight W 1 be added, then the stress on 
the beam when supported at one end will be measured by L x IF ; 
while, in the case of the beam supported at both ends, the weight 
being at the middle point of the beam, the stress is measured as 
before, by J L x W (Art. 191). Therefore, a weight placed at the 
end of a beam supported only at one end produces four times the 
stress as the same weight placed at the middle of the beam when 
supported at both ends. 

1. What must be the length of a beam of oak one inch square, 
supported at both ends, which is just capable of bearing its own 
weight ? 

By Art. 191, a beam of oak 1 foot long and 1 inch square, 

weighing ^ pound, just supports 600 pounds. And by Art. 196, 

the expression I = L I ^ 1 denotes that when a beam 

whose length is L breaks when TF is placed upon it, I is the length 
of a beam that will break with its own weight ; consequently, since 

here L = 1, W = } 2J and TF = 600, I = ft + | 12 ) = (2401)* 

\ y f 

= 49 feet. 

2. Two beams are of equal length and weight, the first being 
a square prism whose section is 4 inches square, the second a rect 
angular prism, 8 by 2 inches ; how much stronger is the second 


beam than the first, and how much stronger when laid on tho nar 
row than on the broad side ? (Art. 180.) 

Ans. The second beam is tmcc as strong as the first, and 
four times as strong when laid on the narrow as 
on the broad side. 

198. Structures Relatively Weaker as they are 
Larger. The foregoing articles explain the observed fact that the 
relative strength of every kind of structure becomes less as its size 
is increased. For, the absolute strength increases as the square 
of one of the dimensions, while the weight increases as the cube of 
the same. A model, therefore, has far greater relative strength 
than the building copied from it ; and in respect to every kind of 
structure, there are limits of magnitude which cannot be exceeded. 

The same fact is observed in the animal and vegetable world. 
Relatively to size, insects are very much stronger than large ani 
mals, and shrubs stronger than trees. 

199. Strength cf Solid and Hollow Cylinders. If a 

solid and a hollow cylinder, of equal length, have the same quan 
tity of matter, so that the area of their cross-sections shall be 
equal, then their strength will be in the ratio cf the distances of 
their centres of gravity from the upper surfaces. But the centres 
of gravity being at the centres of the cross-sections, it follows that 
the strength of the solid cylinder will be less than that of the hol 
low cylinder in the ratio of the diameter of the former to that of 
the latter. 

It appears, therefore, that the strength of a tube is always 
greater than the strength of the same quantity of matter made 
into a solid rod of the same length ; and leaving out of view the 
diminished rigidity, there would seem to be no limit to the strength 
which might be given to such a cylinder by increasing its diameter. 

Many illustrations are found in nature, such as the bones of 
animals, the quills of birds feathers, the straw of grain, and the 
tubular stalks of some larger plants. 

An interesting application of the principle has been made in 
modern times in the construction of iron tubular bridges. 


1. Two forces, F and F 1 , acting in the diagonals of a parallelo 
gram, keep it at rest in such a position that one of its edges is 
horizontal ; show that F sec a = F sec a W cosec (a + a ), 
where W is the weight of the parallelogram, and a and a the 
angles between the diagonals and the horizontal side. 


2. Four parallel forces act at the angles of a plane quadri 
lateral, and are inversely proportional to the segments of its diag 
onals nearest to them ; show that the point of application of their 
resultant lies at the intersection of the diagonals. 

3. Find the centre of gravity of four equal heavy 7 particles 
placed at the four angular points of a triangular pyramid. 

4. Five pieces of a uniform chain are hung at equidistant points 
along a rigid rod without weight, and their lower ends are in a 
straight line passing through one end of the rod; find the centre 
of gravity of the system. 

5. A right square pyramid, whose height is 8 feot, and the 
edge of its base 1 foot, is tipped on one edge till it is on the point 
of falling; what angle does its axis make with the horizon ? 

6. If three uniform rods be rigidly united so as to form half of 
a regular hexagon, prove that if suspended from one of the angles, 
one of the rods will be horizontal. 

7. A cone of uniform density, whose slant height is 15 inches, 
is suspended by the edge of its base, when its axis is found to in 
cline 12 to the horizon ; required the other dimensions of the 

8. If A B C be an isosceles triangle, having a right angle at C, 
and if D and E be the middle points of A C and A B, respect 
ively, prove that a perpendicular from E upon B D will pass 
through the centre of gravity of the triangle B D C. 

9. An oblique cylinder, inclining 62,} to the horizon, having 
slant height = 14 inches, and the diameter of the base 6A 
inches, has a ball of the same material hung upon its edge, which 
just upsets it; required the diameter of the ball. 

10. A body, the lower surface of which is spherical, rests upon 
a horizontal plane ; find in what case the equilibrium is stable, and 
in what case unstable. 

11. A smooth circular ring rests on two pins projecting from a 
wall, and the pins are not in the same horizontal plane ; find the 
pressure on each pin. 

12. A given isosceles triangle is inscribed in a circle ; find the 
centre of gravity of the remaining area of the circle. 

13. A homogeneous hemisphere rests with its convex surface 
on a horizontal plane ; at what points of the circumference of the 
plane base of the hemisphere must three weights of 10, 15, and 20 
Ibs. be suspended, in order that its position be not changed ? 

14. Two smooth cylinders of equal radii just fit in between two 
parallel vertical walls, and rest on a smooth horizontal plane, with 
out pressing against the walls ; if a third equal cylinder be placed 
on the top of them, find the resulting pressure against either 


5, A cylinder, suspended by a point on the side, inclines 40 
to the horizon ; the point is moved 3 feet lengthwise on the side, 
and then the cylinder inclines 24 to the horizon, with the otlu-r 
end down ; find the point of suspension, that the cylinder may- 
hang horizontal. 

16. A flat semicircular board, with its plane vertical, and curved 
edge upward, rests on a smooth horizontal plane, and is pressed at 
two given points of its circumference by two heavy rods, which 
slide freely in vertical guides; find the ratio of the weights of the 
rods, that the board may be in equilibrium. 

17. The radii of a wheel and axle are 12 inches and 3 inches; 
the power is 30 Ibs., the weight 100 Ibs. : as the power in this case 
preponderates, required how many degrees from the bottom of the 
wheel the end of the rope is when the forces are in equilibrium. 

18. A frustum is cut from a right cone by a plane bisecting the 
axis, and parallel to the base; show that it will rest with its slant 
side on a horizontal plane, if the height of the cone have to the 
diameter of its base a greater ratio than Vl to Vl7. 

19. Explain the action of an oar, when used in rowing, and 
determine the effect produced, having given the distances from 
the hands to the side of the boat, and from the side of the boat to 
the point where the oar may be considered as acting on the water. 

20. A uniform wheel, free to revolve on its axis, has the weights, 
21 Ibs. and 13 Ibs., attached to the circumference, 100 apart; how 
far from the bottom will the weights be, respectively, when the 
system is in equilibrium? 

21. Two equal rods, without weight, are connected at their 
middle points by a pin, which allows free motion in a vertical 
plane ; they stand upon a horizontal plane, and their upper ex 
tremities are connected by a thread, which carries a weight. 
Show that the weight will rest half way between the pin and the 
horizontal line joining the upper ends of the rods. 

22. A uniform heavy rod, of given length, is to be supported in 
a given position, with its upper end resting against a smooth ver 
tical wall, by a string fastened to its lower end ; find the point in 
the wall to which the string must be attached. 

23. A light cord, with one end attached to a fixed point, passes 
over a pulley in the same horizontal plane with the fixed point, 
and supports a weight hanging freely at its other end. A heavy 
ring being put upon the cord in different places between the fixed 
point and the pulley, it is required to show that, if the weight of 
the ring be small compared with the other weight, the positions 
of the ring, when in equilibrium, will be approximately in the arc 
of a circle. 

2-1. If particles of unequal weight be placed in the angular 


points of a triangular pyramid, and G be their common centre of 
gravity, G , G", G ", &c., be the common centres of gravity for 
every possible arrangement of the particles ; show that the centre 
of gravity of equal particles, placed at G, G , G", &c., is the centre 
of gravity of the pyramid. 

25. Two equal circular disks with smooth edges, placed on 
their flat sides in the corner between two smooth vertical planes, 
inclined to each other at a given angle, touch each other in the 
line bisecting that angle; find the, radius of the least disk that 
may be pressed between them without causing them to separate. 

26. A ladder of uniform weight throughout, 36 feet long, 
weighs 72 Ibs., and leans against a vertical wall, making an angle 
of 66 40 with the horizon ; a man, weighing 130 Ibs., ascends 30 
feet on the ladder ; required the amount of pressure against the 

27. Where is the centre of gravity of the area included between 
two circles tangent to each other internally ? 





201. Liquids Distinguished from Solids and Gases. 

A fluid is a substance whose particles are moved among each other 
by a very slight force. In solid bodies the particles are held by 
the force of cohesion in fixed relations to each other; hence such 
bodies retain their form in spite of gravity or other small forces 
exerted upon them. If a solid be reduced to the finest powder, 
still each grain of the powder is a solid body, and its atoms are 
held together in a determinate shape. A pulverized solid, if piled 
up, will settle by the force of gravity to a certain inclination, ac 
cording to the smallness and smoothness of its particles, while a 
liquid will not rest till its surface is horizontal. 

Fluids are of two kinds, liquids and gases. In a liquid, there 
is a perceptible cohesion among its particles ; but in a gas, the 
particles mutually repel each other. These fluids are also distin 
guished by the fact that liquids cannot be compressed except in a 
very slight degree, while the gases are very compressible. A force 
of 15 pounds on a square inch, applied to a mass of water, will 
compress it only about .000046 of its volume, as is shown by an 
instrument devised by Oersted. But the same force applied to a 
quantity of air of the usual density at the earth s surface will re 
duce it to one-half of its former volume. 

202. Transmitted Pressure. It is an observed property of 
fluids that a force which is applied to one 

part is transmitted undivided to all parts. 
For instance, if a piston A (Fig. 143) is 
pressed upon the water in the vessel ADO 
with a force of one pound, every other pis 
ton of the same size, as B, C, D, or E, re 
ceives a pressure of one pound in addition 
to the previous pressure of the water itself. 
Hence the whole amount of bursting press 
ure exerted within the vessel by the weight 



upon A equals as many pounds as there are portions of surface 
equal to the area of A. And if the pressure is increased till the 
vessel bursts, the fracture is as likely to occur in some other part 
as in that toward which the force is directed. 

203. The Hydraulic Press. An important application of 
the principle of transmitted pressure occurs in Bramah s hydraulic 
press, represented in Fig. 144. The walls of the cylinder and res- 

FIG. 144. 

ervoir are partly removed, to show the interior. A is a small 
forcing pump, worked by the lever J/, by which water is raised in 
the pipe a from the reservoir H, and driven through the tube K 
into the cylinder B, where it presses up the piston P, and the iron 
plate on the top of it, against the substance above. At each down 
ward stroke of the small piston p, a quantity of water is transferred 
to the cylinder J5, and presses up the large piston with a force as 
many times greater than that exerted on the small one as the 
under surface of P is greater than that of p (Art. 202). If. the 
diameter of p is one inch, and that of P is ten inches, then any 
pressure on p exerts a pressure 100 times as great on P. The lever 
M gives an additional advantage. If the distances from the ful 
crum to the rod p and to the hand are as 1 : 5, this ratio com 
pounded with the other, 1 ; 100, gives the ratio of power at M to 


the pressure at Q as 1 : 500; so that a power of 100 Ibs. exerts a 
pressure of 50000 Ibs. 

This machine has the special advantage of working with a 
small amount of friction. It is used for pressing paper and books, 
packing cotton, hay, &c. ; also for testing the strength of cables 
and steam-boilers. It has been sometimes employed to raise great 
weights, as, for instance, the tubular bridge over the Menai straits ; 
the two portions, after being constructed at the water level, were 
raised more than 100 feet to the top of the piers, by two hydraulic 
presses. The weight of each length lifted at once was more than 
1800 tons. 

The relation of power to weight in the hydraulic press is in 
accordance with the principle of virtual velocities (Art. 142). For, 
while a given quantity of water is transferred from the smaller to 
the larger cylinder, the velocity of the large piston is as much less 
than that of the small one as its area is greater. But we have seen 
that the pressures are directly as the areas. Therefore, in this as 
in other machines, the intensities of the forces are inversely as 
their virtual velocities. 

204 Equilibrium of a Fluid. In order that a fluid may 
be at rest, 

1. The pressures at any one point must be equal in all direc 

2. The surface must be perpendicular to the resultant of the 
forces which act upon it. 

Both of these conditions result from the mobility of the par 
ticles. It is obvious that the first must be true, since, if any 
particle were pressed more in one direction than another, it would 
move in the direction of the greater force, and therefore not be at 
rest, as supposed. 

In order to show the truth of the second condition, let m p 
(Fig. 145) represent the resultant of 
the forces which act on the fluid. Then, FIG. 145. 

if the surface is not perpendicular to 
m p, that force may be resolved into 
m q perpendicular to the surface, and 
m f parallel to it. The latter, mf, not 
being opposed, the particles move in 
that direction. 

As gravity is the principal force which acts on all the particles, 
the surface of a fluid at rest is ordinarily level, that is, perpendicu 
lar to a vertical or plumb line. If the surface is of small extent, 
it is sensibly a plane, though it is really curved, because the verti 
cal lines, to which it is perpendicular, converge toward the centre 
of the earth. 



205. The Curvature of a Liquid Surface. The earth 

being 7912 miles in diameter, a distance of 100 feet on its surface 
subtends an angle of about one second at the centre, and therefore 
the levels of two places 100 feet apart are inclined one second to 
each other. 

The amount of depression for moderate distances is found by 
the formula, d = f L?, in which d is the de 
pression in feet, and L the length of arc in 
miles. Let B E (Fig. 146) be a small arc of 
a great circle on the earth ; then C E is the 
depression. As B E is small, its chord may 
be considered equal to the arc, and B G equal 
to the depression. But B G : B E : : B E : 

B A ; that is, d : L : : L : 7912; or d = - t ^. 

In order to express d in feet, while the other 
lines are in miles, we have 

- J^J*W - L * x 
~ 7912~x5286 " 

This gives, for one mile, d 8 inches; for two miles, d 2 
ft. 8 in.; and for 100 miles, d = GG67 ft, &c. If a canal is 100 
miles long, each end is more than a mile below the tangent to the 
surface of the water at the other end. 

206. The Spirit Level. Since the surface of a liquid at 
rest is level, any straight line which is placed parallel to such a 
surface is also level. Leveling instruments are constructed on 
this principle. The most accurate kind is the one called the 
spirit level. Its most essential -^ 

part is a glass tube, A B (Fig. 
147), nearly filled with alcohol 
(because water would be liable to freeze), and hermetically sealed. 
The tube having a little convexity upward from end to end, 
though so slight as not to be visible, the bubble of air moves to 
the highest part, and changes its place by the least inclination of 
the tube. The tube is so connected with a straight bar of wood 
or metal, as D C (Fig. 148), or for nicer purposes, with a telescope, 
that the bubble is at the 
middle M when the bar 
or the axis of the tele 
scope is exactly level. 
The tube usually has 
graduation lines upon it 
for adjusting the bubble accurately to the middle. 



207. Pressure as Depth. From the principle of equal 
transmission of force in a fluid, it follows that, if a liquid is uni 
formly dense, its pressure on a given area varies as the perpen 
dicular depth, whatever the form or size of the reservoir. Let the 
vessel A B C D (Fig. 149), having the form of a right prism, be 
filled with water, and imagine the water to be divided by horizon 
tal planes into strata of equal thickness. If the density is every 
where the same, the weights of these strata are equal. But the 
pressure on each stratum is the sum of the weights of all tho 
strata above it. Therefore, in this case, the pressure varies as the 

FIG. 149. FIG. 150 FIG. 151 

B / " - A C 19 3 

But let the reservoir A B E H (Fig. 150) contain water which 
is not directly beneath the highest part. The pressures in the 
column A B C D are transmitted laterally to E H, however far 
distant ; so that the surface of each horizontal stratum sustains 
equal pressures in all parts, whether directly beneath A B or not. 
Hence, if G H is equal to C D, the downward pressure on G H is 
equal to the weight of A B C D ; so, also, the upward pressure on 
E F is equal to the weight of A B L M, and would just sustain 
the column of water E F N P. 

Again, if the base is smaller than the top, as in the vessel 
A BE F (Fig. 151), then the pressure on E F equals only the 
weight of the column G D E F. The water in the surrounding 
space ACE, B D F, simply serves as a vertical wall to balance 
the lateral pressures of the central column. 

If the surface pressed upon is oblique or vertical, then the 
points of it are at unequal depths ; in this case, the depth of the 
area is understood to be the average depth of all its parts ; that 
is, the depth of its centre of gravity. 

If the fluid were compressible, the lower strata would be moro 
dense than the upper ones, and therefore the pressure would in 
crease at a faster rate than the depth. 

208. Amount of Pressure in Water. One cubic foot of 
water weighs 1000 ounces, or 62.5 pounds. Therefore, the pressure 
on one square foot, at the depth of one foot, is G2.5 pounds. !> MU 
this, as the unit of hydrostatic pressure, it is easy to detcrmiu- 



pressures on all surfaces, at all depths ; for it is obvious that, when 
the depth is the same, the pressure varies as the surface pressed 
upon ; and it has been shown that, on a given surface, the press 
ure varies as the depth of its centre of gravity ; it therefore varies 
as the product of the two. Let p pressure ; a = area pressed 
upon; and d = the depth of its centre of gravity; then 
p = a d x 62.5. 

Depth. Pounds per sq. ft. Depth. Pounds per sq. ft. 

I ft 62.5 

10 625 

16 . . .1000 

100 ft 6,250 

i mile 330,000 

5 miles 1,650,000 

FIG. 152. 

From the above table it may be inferred that the pressure on 
a square foot in the deepest parts of the ocean must be not far 
from two millions of pounds; for the depth in some places is 
more than five miles, and sea-water weighs 64.37 pounds, instead 
of 62.5 pounds. A brass vessel full of air, containing only a pint, 
and whose walls were one inch thick, has been known to be 
crushed in by this great pressure, when 
sunk to the bottom of the ocean. 

Owing to the increase of pressure 
with depth, there is great difficulty in 
confining a high column of water by 
artificial structures. The strength of 
banks, dams, flood-gates, and aqueduct 
pipes, must increase in the same ratio 
as the perpendicular depth from the sur 
face of the water, without regard to its 
horizontal extent. 

209. Column of Water whose 
Weight Equals the Pressure. A 

convenient mode of conceiving readily 
of the amount of pressure on an area, 
in any given circumstances, is this: 
consider the area pressed upon to form 
the horizontal base of a hollow prism ; 
let the height of the prism equal the 
average depth of the area; and then 
suppose it filled with water. The weight 
of this column of water is equal to the 
pressure. For the contents of the prism 
(whose base = a, and its height d), 
= ad , and the weight of the same = 
a d x 62.5 Ibs. ; which is the same ex 
pression as was obtained above for the 



On the bottom of a cubical vessel full of water, the pressure 
equals the weight of the water; on each side of the same the 
pressure is one-half the weight of the water ; hence, on all the five 
sides the pressure is three times the weight of the water ; and if 
the top were closed, on which the pressure is zero, the pressure on 
the six sides is the same, three times the weight of the water. 

210. Illustrations of Hydrostatic Pressure. A vessel 
may be formed so that both its base and height shall be great, but 
its cubical contents small ; in which case, a great pressure is pro 
duced by a small quantity of water. The hydrostatic bellows is 
an example. In Fig. 152, the weight which can be sustained on 
the lid D I by the column A D is equal to that of a prism or cyl 
inder of water, whose base is D I, and its height D A. It is im 
material how shallow is the stratum of water on the base, or how 
slender the tube A D, if greater than a capillary size. 

In like manner, a cask, after being filled, may be burst by an 
additional pint of water ; for, by screwing a long and slender pipe 
into the top of the cask, and filling it with water, the pressure is 
easily made greater than the strength of the cask can bear. 

211. The Same Level in Connected Vessels. In tubes 
or reservoirs which communicate with each other, water will rest 
only when its surface is at the 

same level in them all. If water 
is poured into D (Fig. 153), it 
will rise in the vertical tube B, 
so as to stand at the same level 
as in D. For, the pressure to 
ward the right on any cross-sec 
tion E of the horizontal pipe 
m n equals the product of its 
area by its depth below D. So 
the pressure on the same section 
towards the left equals the pro 
duct of its area by its depth be 
low B. But these pressures are 
equal, since the liquid is at 

rest. Therefore E is at equal depths below B and D ; in other 
words, B and D are on the same level. The same reasoning ap 
plies to the irregular tubes A and (7, and to any others, of what 
ever form or size. 

Water conveyed in aqueducts, or running in natural channels 
in the earth, will rise just as high as the source, but no higher. 

Aiirt/au ivells illustrate the same tendency of water to rise to 
its level in the different branches of a tube. When a deep boring 

FIG 153. 



is made in the earth, it may strike a layer or channel of water 
which descends from elevated land, sometimes very distant. The 
pressure causes it to rise in the tube, and often throws it manv 
feet above the surface. Fig. 154 shows an artesian well, through 
which is discharged the water that descends in the porous stratum 
K K, confined between the strata of clay A B and C D. 

FIG. 154. 

212. Centre of Pressure. The centre of pressure of any 
surface immersed in water is that point through which passes the 
resultant of all the pressures on the surface. It is the point, 
therefore, at which a single force must be applied in order to 
counterbalance all the pressures exerted on the surface. If the 
surface be a plane, and horizontal, the centre of pressure coincides 
with the centre of gravity, because the pressures are equal on every 
part of it, just as the force of gravity is. But if the plane surface 
makes an angle with the horizon, the centre of pressure is lower 
than the centre of gravity, since the pressure increases with the 
depth. For example, if the vertical side of a vessel full of water 
i^ rectangular, the centre is one-third of the distance from the 
middle of the base to the middle of the upper side. If triangular, 
with its base horizontal, the centre of pressure is one-fourth of the 
distance from the middle of the base to the vertex. If triangular, 
with the top horizontal, the centre of pressure is half way up on 
the bisecting line. 

[See Appendix for calculations of the place of the centre of 

213. The Loss of Weight in Water. When a body is 
immersed in water, it suffers a pressure on every side, which is 
proportional to the depth. Opposite components of lateral press 
ures, being exerted on surfaces at the same depth, balance each 
other ; but this cannot be true of the vertical pressures, since the 
top and bottom of the body are at unequal depths. The upward 
pressure on the bottom exceeds the downward pressure on the top ; 


and this excess constitutes the buoyant power of a fluid, which 
causes a loss of weight. 

A body immersed in water loses weight equal to the weight of 
water displaced. 

For before the body was immersed, the water occupying the 
same space was exactly supported, being pressed upward more 
than downward by a force equal to its own weight. The weight 
of the body, therefore, is diminished by this same difference of 
pressures, that is, by the weight of the displaced water. 

On the supposition of the complete incompressibility of water, 
this loss is the same at all depths, because the weight of displaced 
water is the same. As water, however, is slightly compressible, its 
buoyant power must increase a little at great depths. Calling the 
compression .000046 for one atmosphere (=34 feet of water), the 
bulk of water at the depth of a mile is reduced by about T J ^, and 
its specific gravity increased in the same ratio ; so that, possibly, a 
body might sink near the surface, and float at great depths in the 
ocean. But this is not probable in any case, since the same com 
pressing force may reduce the volume of the solid as much as that 
of the water. And, furthermore, the increase of density by in 
creased depth is so slow, that even if solids were incompressible, 
most of those which sink at all would not find their floating placo 
within the greatest depths of the ocean. For example, a stone 
twice as heavy as water must sink 100 miles before it could float. 

214. Equilibrium of Floating Bodies. If the body which 
is immersed has the same density as water, it simply loses its 
whole weight, and remains wherever it is placed. But if it is less 
dense than water, the excess of upward pressure is more than suf 
ficient to support it; it is, therefore, raised to the surface, and 
comes to a state of equilibrium after partly emerging. In order 
that a floating body may have a stable equilibrium, the three fol 
lowing conditions must be fulfilled : 

1. It displaces an amount of water whose weight is equal to its 

2. Tlie centre of gravity of the lody is in the same vertical line 
ivith that of the displaced water. 

3. The metacentcr is higher than the centre of gravity of the 

The reason for the first condition is obvious; for both the body 
and the water displaced by it are sustained by the same upward 
pressures, and therefore must be of equal weight. 

That the second is true, is proved as follows : Let C (Fig. 155, 1) 
be the centre of gravity of the displaced water, while that of the 
body is at G. Now the fluid, previous to its removal, was ur> 



tained by an upward force equal to its own weight, acting through 
its centre of gravity (?; and the same upward force now acts upon 

FIG. 155. 

the floating body through the same point. But the body is urged 
downward by gravity in the direction of the vertical line AGE. 
Were these two forces exactly opposite and equal, they would keep 
the body at rest; but this is the case only when the points C and 
G are in the same vertical line : in every other position of these 
points, the two parallel forces tend to turn the body round on a 
point between them. 

215. The Metacenter. To understand the third condition, 
the metacenter must be defined. When a floating body is slightly 
inclined from its state of equilibrium, as in Fig. 155, 2 and 3, and 
a vertical is drawn through the new centre of gravity C of the dis 
placed water, this vertical must intersect the former vertical A J9; 
the intersection, J/, is called the metacenter. When the centre of 
gravity of the body G is lower than the metacenter, as in Fig. 155, 2 9 
the parallel forces, downward through G and upward through C\ 
revolve the body back to its position of equilibrium, which is then 
called a stable equilibrium. But if the centre of gravity of the 
body is higher than the metacenter, as in Fig. 155, 3, the rotation 
is in the opposite direction, and the body is upset, the equilibrium 
being unstable. Once more, if the centre of gravity of the body is 
at the metacenter, the body rests indifferently in any position, as, 
for example, a sphere of uniform density. The equilibrium in this 
case is called neutral. 

If only the first condition is fulfilled, there is no equilibrium ; 
if only the first and second, the equilibrium is unstable; if all the 
three, the equilibrium is stable. 

In accordance with the third condition, it is necessary to place 
the heaviest parts of a ship s cargo in the bottom of the vessel, and 
sometimes, if the cargo consists of light materials, to fill the bot 
tom with stone or iron, called ballast, lest the masts and rigging 
should raise the centre of gravity too high for stability. On the 
same principle, those articles which are prepared for-life-preservers, 


in case of shipwreck, should be attached to the upper part of UK- 
body, that the head may be kept above water. The danger an 
from several persons standing up in a small boat is quite apparent ; 
for the centre of gravity is elevated, and liable to become higher 
than the metacenter, thus producing an unstable equilibrium. 

216. Floating in a Small Quantity of Water, As press 
ure on a given surface depends solely on the depth, and not at all 
on the extent or quantity of water, it follows that a body will float 
as freely in a space slightly larger than itself as on the open water 
of a lake. For instance, a ship may be floated by a few hogsheads 
of water in a dock whose form is adapted to it. In such a cas-?, it 
cannot be literally true that the displaced water weighs as much 
as the vessel, when all the water in the dock may not weigh a 
hundredth part as much. The expression " displaced water " means 
the amount which would fill the place occupied by the immersed 
portion of the body. An experiment illustrative of the above is, to 
float a tumbler within another by means of a spoonful of water 

217. Floating of Heavy Substances. A body of the 

most dense material may float, if it has such a form given it as to 
exclude the water from the upper side, till the required amount is 
displaced. Ships are built of iron, and laden with substances of 
greater specific gravity than water, and yet ride safely on the ocean. 
A block of any heavy material, as lead, may be sustained by the 
upward pressure beneath it, provided the water is excluded from 
the upper side by a tube fitted to it by a y/ater-tight joint. 

218. Specific Gravity. The weight of a body compared 
with the weight of the same volume of the standard, is called its 
specific gravity. 

Distilled water, at about 39 F., the temperature of its greatest 
density, is the standard for ail solids and liquids, and common air, 
at 32, for gases. Therefore the specific gravity of a solid or a 
liquid body, is the ratio of its weight to the weight of an equal 
volume of water ; and the specific gravity of an aeriform body is 
the ratio of its weight to the weight of an equal volume of air. 
Hence, to find the specific gravity of a solid or liquid, divide its 
weight by the weight of the same volume of water ; but in the case 
of a gas, divide by the weight of the same volume of air. 

219. Methods of Finding Specific Gravity. 

1. For a solid heavier than water, divide its weight by its lorn 
of weight in water. 

The reason for this rule is obvious. Tho weight which a sub 
merged body loses (Art. 213) is equal to the weight of the dis- 



placed water, which has, of course, the same volume as the body ; 
therefore, dividing by the loss is the same as dividing by the V/eight 
of the same volume of water. 

2. For a solid lighter than water, divide its weight by its weight 
added to the loss it occasions to a heavier body previously balanced 
in ivater. 

For, if the light body be attached to a body heavy enough to 
sink it, it loses all its own weight, and causes loss to the other 
which was previously balanced. And the whole loss equals the 
weight of water displaced by the light body. Hence, as before, we 
in fact divide the weight of the body by the weight of the same 
volume of water. 

3. For a liquid, find the loss which a body sustains weighed in 
the liquid and then in water, and divide the first loss by the second. 

For the first loss equals the weight of the displaced liquid, and 
the second that of the displaced water; and the volume in each 
case is the same, namely, that of the body weighed in them. 

But the specific gravity of a liquid may be more directly ob 
tained by measuring equal volumes of it and of water in a flask, 
and finding the weight of each. Then the weight of the liquid 
divided by that of the water is the specific gravity required. 

220. The Hydrometer, or Areometer. In commerce and 
the arts, the specific gravities of substances are obtained in a more 
direct and sufficiently accurate way, by instruments constructed 
for the purpose. The general name for such instruments is the 
hydrometer, or areometer. But other names are given to such as 
are limited to particular uses ; as, for example, the alcoometer for 
alcohol, and the lactometer for milk. The hydrometer, represented 
in Fig. 156, consists of a hollow ball, with a 
graduated stem. Below the ball is a bulb con 
taining mercury, which gives the instrument a 
stable equilibrium when in an upright position. 
Since it will descend until it has displaced a 
quantity of the fluid equal in weight to itself, it 
will of course sink to a greater depth if the fluid 
is lighter. From the depths to which it sinks, 
therefore, as indicated by the graduated stem, 
the corresponding specific gravities are esti 

Nicholson s hydrometer (Fig. 157) is the most 
useful of this class of instruments, since it may 
be applied to finding the specific gravities of 
solid as well as liquid bodies. In addition to 
the hollow ball of the common hydrometer, it is furnished at the 



FIG. 157. 

top with a pan A for receiving weights, and a cavity beneath for 
holding the substance under trial. The instrument is so adjusted 
that when 1000 grains are placed in the pan, the instrument sinks 
in distilled water at the temperature of 
39 F. to a fixed mark, 0, on the stem. 
Calling the weight of the instrument W, 
the weight of displaced water is W 4- 1000. 
To find the specific gravity of a liquid, 
place in the pan such a weight zv as will 
just bring the mark to the surface. Then 
the weight of the liquid displaced is TT-f w. 
But its volume is equal to that of the dis 
placed water. Therefore its specific grav- 
W + w 

To find the specific gravity of a solid, 
place in the pan a fragment of it weighing 
less than 1000 grains, and add the weight 
w required to sink the mark to the water- 
level. Then the weight of the substance 
in air is 1000 20. Remove the substance 
to the cavity at the bottom of the instrument, and add to the 
weight in the pan a sufficient number of grains w to sink the 
mark to the surface. Then w is the loss of weight in water ; 

therefore, - -, is the specific gravity of the substance. 

221. Specific Gravity of Liquids by Means of 

Heights. The specific gravity of two liquids may be compared 
by their relative heights when in equilibrium. Let the tubes m 
and n (Fig. 158) communicate with 
each other, and be furnished with 
a scale of heights above the zero 
line B C. Suppose the column of 
water A B to be in equilibrium 
with the column of mercury CD. 
Put h = the height of the water, 
h = the height of the mercury, and 
s = the specific gravity of the lat 
ter; then, since pressure varies as 
the product of height and density, 
and the pressures in this case are 
equal, we have h x 1 = h x s; 

whence s = j r ; that is, the specific 

FIG. 158. 


gravity is found by dividing the height of the water by the height 
of the liquid. Also, h .h :: s : 1; that is, the heights of two 
columns in equilibrium are inversely as their specific gravities. 

The heavier liquid should be poured in first, till it stands 
somewhat above B C y the zero mark of the scale ; and then the 
lighter should be poured into one branch, till it presses the other 
down to the zero line. The heights of both are reckoned upward 
from B C, since the heavy liquid below B G balances itself. 

222. Table of Specific Gravities. An accurate knowl 
edge of the specific gravities of bodies is important for many pur 
poses of science and art, and they have therefore been determined 
with the greatest possible precision. The heaviest of all known 
substances is platinum, whose specific gravity, when compressed 
by rolling, is 22, water being 1 ; and the lightest is hydrogen, whose 
specific gravity is .073, common air being 1. Now, as water is 
about 800 times as heavy as air, it is (800 -r .073 =) 10,959 times 
as heavy as hydrogen. Therefore platinum is about (10,959 x 22 = ) 
241,000 times as heavy as hydrogen. Between these limits, 1 and 
241,000, there is a wide range for the specific gravities of all other 
substances. As a class, the common metals are the heaviest 
bodies; next to these come the metallic ores; then the precious 
gems ; minerals in general, animal and vegetable substances, as 
shown in the following table ; 

Metals (pure), not including the bases of the alkalies and 

earths, from 5 22 

Platinum .... 22.0 

Gold 19.25 

Mercury 13.58 

Lead 11.35 

Silver 10.47 

Copper 8.90 

Steel 7.84 

Iron 7.78 

Tin . . 7.29 

Zinc 7.00 

Metallic ores, lighter than the pure metals, but usually 

above 4.00 

Precious gem s, as the ruby, sapphire, and diamond . . . 3 4 

Minerals, comprehending most stony bodies 2 3 

Liqidds, from ether highly rectified to sulphuric acid highly 

concentrated f 2 

Acids in general, heavier than water. 
Oils in general, lighter ; but the oils of cloves and cinna 
mon are heavier than water ; the greater part lie between 

.9 and i 9 I 

Milk. . 1.032 

Alcohol (perfectly pure) 797 

" of commerce 835 

Proof spirit 923 

Wines ; the specific gravity of the lighter wines, as Cham 
pagne and Burgundy, is a little less, and of the heavier 
wines, as Malaga, a little greater than that of water. 
Woods, cork being the lightest, and lignum vitse the heaviest .24 1.34 


223. Floating. The human body, when the lungs are filled 
with air, is lighter than water, and but for the difficulty of keeping 
the lungs constantly inflated, it would naturally float. With a mod 
erate degree of skill, therefore, swimming becpmes a very easy pro 
cess, especially in salt water. When, however, a man plunges, as 
divers sometimes do, to a great depth, the air in the lungs becomes 
compressed, and the body does not rise except by muscular effort. 
The bodies of drowned persons rise and float after a few days, in 
consequence of the inflation occasioned by putrefaction. 

As rocks are generally not much more than twice as heavy as 
water, nearly half their w r eight is sustained while they are under 
water ; hence, their weight seems to be greatly increased as soon 
as they are raised above the surface. It is in part owing to their 
diminished weight that large masses of rock are transported with 
great facility by a torrent. While bathing, a person s limbs feel as 
if they had nearly lost their weight, and when he leaves the water, 
they seem unusually heavy. 

224. To Find the Magnitude of an Irregular Body. 

It would be a long and difficult operation to find the exact con 
tents of an irregular mineral by direct measurement. But it 
might be found with facility and accuracy by weighing it in air 
and then finding its loss of weight in water. The loss is the weight 
of a mass of water having the same volume. Now, as 1000 ounces 
of water measure 1728 cubic inches, a direct proportion will show 
what is the volume of the displaced water ; that is, of the mineral 

225. Cohesion and Adhesion. What distinguishes a 
liquid from a solid is not its want of cohesion so much as the 
mobility of its particles. It is proved in many ways that the par 
ticles of a liquid strongly attract each other. It is owing to this 
that water so readily forms itself into drops. The same property 
is still more observable in mercury, which, when minutely divided, 
will roll over surfaces in spherical forms. When a disk of almost 
any substance is laid upon water, and then raised gently, it lifts a 
column of water after it by adhesion, till at length the edge of the 
fluid begins to divide, and the column is detached, not in all parts 
at once, but by a successive rupturing of the lateral surface. It is 
proved that the whole attraction of the liquid would be far too 
great to be overcome by the force applied to pull off the disk, were 
it not that it is encountered by little and little, at the edges of the 
column. But it is the cohesion of the water which is overcome in 
this experiment ; for the upper lamina still adheres to the disk. 
By a pair of scales we find that it requires the same force to draw 
off disks of a given size, whatever the materials may be, provided 



they are wet when detached. This is what might be expected, 
since in each case we break the attraction between two laminae of 
water. But if we use disks which are not wet by the liquid, it is 
not generally true that those of different material will be removed 
by the same force ; indicating that some substances adhere to a 
given liquid more strongly than others. 

These molecular attractions extend to an exceedingly small 
distance, as is proved by many facts. A lamina of water adheres 
as strongly to the thinnest disk that can be used as to a thick one ; 
so, also, the upper lamina coheres with equal force to the next 
below it, whether the layer be deep or shallow. 

226. Capillary Action. This name is given to the molecular 
forces, adhesion and cohesion, when they produce disturbing 
effects on the surface of a liquid, elevating it above or depressing 
it below the general level. These effects are called capillary, be 
cause most strikingly exhibited in very fine (hair-sized) tubes. 

The liquid will be elevated in a concave curve, or depressed in a 
convex curve, ly the side of the solid, according as the attraction of 
the liquid molecules for each other is less or greater than twice the 
attraction between the liquid and the solid. 

Case 1st. Let H K (Fig. 159, 1) and L M be a section of the 
vertical side of a solid, and of the general level of the liquid. The 


particle A, where these lines meet, is attracted (so far as this sec 
tion is concerned) by all the particles of an insensibly small quad 
rant of the liquid, the resultant of which attractions is in the line 
A D, 45 below A M. It is also attracted by all the particles in 
two quadrants of the solid, and the resultants are in the directions 
A B 9 45 above, and A B , 45 below L M. 

Now suppose the force A D to be less than twice A B or A B r . 
Cut off C D = A B ; then A B, being opposite and equal to C D, 
is in equilibrium with it. The remainder A C, being less than 
A B , their resultant A E will be directed toward the solid ; and 
therefore the surface of the liquid, since it must be perpendicular 


to the resultant of forces acting on it (Art. 204), takes the direc 
tion represented ;. that is, concave upward. 

Case 2d. Let A D (Fig. 159, 2), the attraction of A toward 
the liquid particles, be more than twice A B, the attraction toward 
a quadrant of the solid. Then, making G D equal to A B, these 
two resultants balance as before; and as A C is greater than A B , 
the angle between A C and the resultant A E is less than 45, and 
A is drawn away from the solid. Therefore the surface, being 
perpendicular to the resultant of the molecular forces acting on it, 
is convex upward. 

Case 3d. If A D (Fig. 159, 3) be exactly twice A B, then CD 
balances A B, and the resultant of A C and A B is A E in a ver 
tical direction ; therefore the surface at A is level, being neither 
elevated nor depressed. 

Case 1st occurs whenever a liquid readily wets a solid, if 
brought in contact with it, as, for example, water and clean glass. 
Case 2d occurs when a solid cannot be wet by a liquid, as glass and 
mercury. Case 3d is rare, and occurs at the limit between the 
other two ; water and steel afford as good an example as any. 

227. Capillary Tubes. In fine tubes these molecular forces 
affect the entire columns as well as their edges. If the material 
of the tube can be wet by a liquid, it will raise a column of that 
liquid above the level, at the same time making the top of the 
column concave. If it is not capable of being wet, the liquid is 
depressed, and the top of the column is convex. The first case is 
illustrated by glass and water ; the second by glass and mercury. 

The materials being given, the distance by which the liquid is 
elevated or depressed varies inversely as the diameter. Therefore 
the product of the two is constant 

The amount of elevation and depression depends on the 
strength of the molecular forces, rather than on the specific 
gravity of the liquids. Alcohol, though lighter than water, is 
raised only half as high in a glass tube. 

If the upper part of a tube is capillary, while the lower part is 
large, a liquid is sustained (after being raised by suction) at the 
same height as if the whole were capillary. But it is found that 
the large mass in the lower part is upheld by atmospheric pressure 
after the capillary part has been closed by the molecular attraction. 

228. Parallel and Inclined Plates. Between parallel 
plates a liquid rises or falls half as far as in a tube of the same 
diameter. This is because the sustaining force acts only on two 
sides of each filament, while in a tube it acts on all sides. There 
fore, as in tubes the height varies inversely as the diameter, so in 
plates the height varies inversely as the distance between them. 




If the plates are inclined to each other, having their edge of 
meeting perpendicular to the horizon, the surface of a liquid rising 
between them assumes the form of a hyperbola, whose branches 
approach the vertical edge, and the water-level, as the asymptotes 
of the curve. This results from the law already stated, that the 
height varies inversely as the distance between the plates. Let 
the edge of meeting, A //(Fig. 160), 
be the axis of ordinates, and the 
line in which the level surface of 
the water intersects the glass, A P, 
the axis of abscissas. Let B C, 
D E, be any ordinates, and A B, 
A D, their abscissas, and B L, D K, 
the distances between the plates. 
By the law of capillarity, the heights 
BC,DE, are inversely &sBL,D K. 
But, by the similar triangles, ABL, 

therefore, B C, D E, are inversely as A B, A D ; and this is a 
property of the hyperbola with reference to the centre and asymp 
totes, that the ordinates are inversely as the abscissas. 

229. Effects cf Capillarity on Floating Bodies. Some 
cases of apparent attractions and repulsions between floating 
bodies are caused by the forms which the liquid assumes on the 
sides of the bodies. If two balls raise the water about them, and 
are so near to each other that the concave surfaces between them 
meet in one, they immediately approach each other till they touch; 
and then, if either be moved, the other will follow it. The water, 
which is raised and hangs suspended between them, draws them 

Again, if each ball depresses the water around it, they will also 
move to each other, and be held together, so soon as they are near 
enough for the convex surfaces to meet. In this case, they are not 
pulled, but pushed together by the hydrostatic pressure of the 
higher water on the outside. 

Once more, if one ball raises the water, and the other depresses 
it, and they are brought so near each other that the curves meet, 
they immediately move apart, as if repelled. For now the equi 
librium is destroyed in a way just the reverse of the preceding 
cases. The water between the balls is too high for that which de 
presses, and too low for that which raises the water, so that the 
former is pushed away, and the latter is drawn away. 

The first case, which is by far the most common, explains the 
fact often observed, that floating fragments are liable to be gath- 


ered into clusters ; for most substances are capable of being wet, 
and therefore they raise the water about them. 

230. Illustrations of Capillary Action. It is by capil 
lary action that a part of the water which falls on the earth is 
kept near its surface, instead of sinking to the lowest depths of 
the soil. This force aids the ascent of sap in the pores of plants. 
It lifts the oil between the fibres of the lamp- wick to the place of 
combustion. Cloth rapidly imbibes moisture by its numerous 
capillary spaces, so that it can be used for wiping things dry. If 
paper is not sized, it also imbibes moisture quickly, and can be 
used as blotting-paper ; but when its pores are filled with sizing, 
to fit it for writing, it absorbs moisture only in a slight degree, 
and the ink which is applied to it must dry by evaporation. 

The great strength of the capillary force is shown in the effects 
produced by the swelling of wood and other substances when kept 
wet. Dry wooden wedges, driven into a groove cut around a 
cylinder of stone, and then occasionally wet, will at length cause 
it to break asunder. As the pores between the fibres of a rope 
run around it in spiral lines, the swelling of a rope caused by 
keeping it wet will contract its length with immense force. 

231. Questions in Hydrostatics. 

1. The diameters of the two cylinders of a hydraulic press are 
one inch and one foot, respectively; before the piston descends, the 
column of water in the small cylinder is two feet higher than the 
bottom of the large piston. Suppose that by a screw a force of 
500 Ibs. is applied to the .small piston ; what is the whole force ex 
erted on the large piston at the beginning of the stroke ? 

Ans. 7209817 Ibs. 

2. A junk bottle, whose lateral surface contained 50 square 
inches, being let down into the sea 3000 feet, what pressure do 
the sides of the bottle sustain, no allowance being made for 
the increased specific gravity of the sea- water ? 

Ans. 65104.166 Ibs. 

3. A Greenland whale sometimes has a surface of 3600 square 
feet ; what pressure would he bear at the depth of 800 fathoms ? 

Ans. 1080,000,000 Ibs., or more than 482,142 tons. 

4. A mill-dam, running perpendicularly across a river, slopes 
at an angle of 25 degrees with the horizon. The average depth 
of the stream is 12 feet, and its breadth 500 yards; required the 
amount of pressure on the dam ? 

Ans. 15,971,906 Ibs., or 7130 + tons. 

5. A mineral weighs 960 grains in air, and 739 grains in water; 
what is its specific gravity ? Ans. 4.344. 


6. What are the respective weights of two equal cubical masses 
of gold and cork, each measuring 2 feet on its linear, edge ? 

Ans. The gold weighs 9625 Ibs. = 4.297 tons; the cork 
weighs 120 Ibs. 

7. A mass of granite contains 5949 cubic feet. The specific 
gravity of a fragment of it is found to be 2.6 ; what does the mass 
weigh ? Ans. 431.568 tons. 

8. An island of ice rises 30 feet out of water, and its upper 
surface is a circular plane, containing fths of an acre. On the 
supposition that the mass is cylindrical, required its weight, and 
depth below the water, the specific gravity of sea-water being 
1.0263, and that of ice .92. 

Ans. Weight, 242,900 tons; depth, 259.64 feet. 

9. Wishing to ascertain the exact number of cubic inches in a 
very irregular fragment of stone, I ascertained its loss of weight in 
water to be 5.346 ounces ; required its volume. 

Ans. 9.238 cubic inches. 

10. Hiero, king of Syracuse, ordered his jeweller to make him 
a crown of gold containing 63 ounces. The artist attempted a 
fraud by substituting a certain portion of silver; which being sus 
pected, the king appointed Archimedes to examine it. Archi 
medes, putting it into water, found it raised the fluid 8.2245 
inches; and having found that the inch of gold weighs 10.36 
ounces, and that of silver 5.85 ounces, he discovered what part of 
the king s gold had been purloined ; it is required to repeat the 
process. Ans. 28.8 ounces. 

11. The specific gravity of lead being 11.35; of cork, .24; of 
fir, .45 ; how much cork must be added to 60 Ibs. of lead, that the 
united mass may weigh as much as an equal bulk of fir ? 

Ans. 65.8527 Ibs. 

12. A cone, whose specific gravity is |, floats on water with its 
vertex downward ; what part of the axis is immersed ? 

Ans. One-half. 

13. A cone, having the same specific gravity as the above, 
floats with its vertex upward ; how much of its axis is immersed ? 

Ans. 0.0436. 

. 14. What is the weight of a chain of pure gold; which raises 
the water 1 inch in height, in a cubical vessel whose side is 3 
inches ? and suppose a chain of the same weight were adulterated 
with 14^ ounces of silver ; how much higher would it raise the 
water in the vessel ? 1 ft. water == 911.458 oz. troy. 

Ans. Weight = 91.35 oz. ; height .133 in. more. 




232. Depth and Velocity of Discharge. From an aper 
ture which is small, compared with the breadth of the reservoir, 
the velocity of discharge varies as the square root of the depth. For 
the pressure on a given area varies as the depth (Art. 207). If the 
area is removed, this pressure is a force which is measured by the 
momentum of the water ; therefore the momentum varies as the 
depth (d). But momentum varies as the mass (q) multiplied by 
the velocity (v) ; hence q v oc d. But it is obvious that q and v 
vary alike, since the greater the velocity, the greater in the same 

ratio is the quantity discharged. Therefore, ( x d, or q x d~ ; 

also v* oc d, or v x d s . 

Not only does the velocity vary as the square root of the depth 
of the orifice, but it is equal to that acquired ~by a lody falling 
through the depth. 

Let h the height of the liquid above the orifice, and h = the 
height of an infinitely thin layer at the orifice. 

If this thin layer were to fall through the height h , under the 
action of its own weight or pressure, the velocity acquired would be 
v = Vtgh (Art. 28). 

Denoting the velocity generated by the pressure of the entire 
column by v, we have, since velocity x 1/depth, 


v : </Zgh : : Vh : \ r h : : 
.-. v = VWgli. 

But VZgh is also the velocity acquired in falling through the 
distance h (Art. 28). 

From an orifice 16 T V feet below the surface of water, the veloc 
ity of discharge is 32 feet per second, because this is the velocity 
acquired in falling 16^ feet; and at a depth four times as great, 
that is, 64 J feet, the velocity will only be doubled, that is, 64] feet 
per second. 

As the velocity of discharge at any depth is equal to that of a 
body which has fallen a distance equal to the depth, it is theoreti 
cally immaterial whether water is taken upon a wheel from a gate 
at the same level, or allowed to fall on the wheel from the top of 
the reseiToir. In practice, however, the former is best, on ac- 


count of the resistance which water meets with in falling through 
the air. 

233. Descent of Surface. When water is discharged from 
the bottom of a cylindric or prismatic vessel, the surface descends 
with a uniformly retarded motion. For the velocity with which 
the surface descends varies as the velocity of the stream, and there 
fore as the square root of the depth (Art. 232). But this is a 
characteristic of uniformly retarded motion, that it varies as the 
square root of the distance from the point where the motion ter 
minates, as in the case of a body ascending perpendicularly from 
the earth. 

The descent of the surface of water in a prismatic vessel has 
been used for measuring time. The clepsydra, or water-clock of 
the Eomans, was a time-keeper of this description. The gradua 
tion must increase upward, as the odd numbers 1, 3, 5, 7, &c. ; 
since, by the law of this kind of motion, the spaces passed over in 
equal times are as those numbers. 

If a prismatic vessel is kept full, it discharges tivice as much 
water in the same time as if it is allowed to empty itself. For the 
.velocity, in the first instance, is uniform; and in the second it is 
uniformly retarded, till it becomes zero. We reason in this case, 
therefore, as in regard to bodies moving uniformly, and with mo 
tion uniformly accelerated from rest, or uniformly retarded till it 
ceases (Art. 25), that the former motion is twice as great as the 

234. Discharge from Orifices in Different Situations. 

Other circumstances besides area and depth of the aperture are 
found to have considerable influence on the velocity of discharge. 
Observations on the directions of the filaments are made by intro 
ducing into the water particles of some opaque substance, having 
the same density as water, whose movements are visible. From 
such observations it appears that the particles of water descend in 
vertical lines, until they arrive within three or four inches of the 
aperture, when they gradually turn in a direction more or less 
oblique toward the place of discharge. This convergence of the 
filaments extends outside of the vessel, and causes the stream to 
diminish for a short distance, and then increase. The smallest 
section of the stream, called the vena contracta, is at a distance 
from the aperture varying from one-half of its diameter to the 

If water is discharged through a circular aperture in a thin 
plate in the bottom of the reservoir, and at a distance from the 
sides, as in Fig. 161, 1, the filaments form the vena contracta at a 
distance beyond the aperture equal to one-half of its diameter; the 



area of the section at the vena contrncta is less than two-thirds 
(0.64) of the area of the aperture ; and the quantity discharged is 
also about two-thirds of that obtained by calculation for the full 
size of the aperture. 

FIG. 161. 

If the reservoir terminates in a short pipe or ajutage, whose 
interior is adapted to the curvature of the filaments, as far as to 
the vena contracta, or a little beyond, as in Fig. 161, 2, it is found 
the most favorable for free discharge, which in some cases reaches 
0.98 of the theoretical discharge. The stream is smooth and pel 
lucid like a rod of glass. The most unfavorable form is that in 
which the ajutage, instead of being external, as in the case just 
described, projects inward, as in Fig. 161, 3 ; the filaments in this 
case reach the aperture, some ascending, others descending, and 
therefore interfere with each other. Hence the stream is much 
roughened in its appearance, and the flow is only 0.53 of what is 
due to the size of the aperture and its depth. 

"When the aperture is through a thin plate, the contraction of 
the stream and the amount of discharge are both modified by the 
circumstance of being near one or more sides of the reservoir. 
There is little or no contraction on the side next the wall of the 
vessel, since the filaments have no obliquity on that side ; and the 
quantity is on that account increased. The filaments from the 
opposite side also divert the stream a few degrees from the perpen 
dicular (Fig. 161, 4). 

235. Friction in Pipes. As has. just been stated, an ajutage 
extending to or slightly beyond the vena contracta, and adapted 
to the form of the stream, very much increases the quantity dis 
charged ; but beyond that, the longer the pipe, the more does it 
impede the discharge by friction. The friction varies directly as 
the length of the pipe, and inversely as its diameter. In order, 
therefore, to convey water at a given rate through a long pipe, it is 
necessary either to increase the head of water or to enlarge the 
pipe, so as to compensate for friction. If a given quantity of 



water is discharged per second by an aperture five inches in diam 
eter, through the thin bottom of a reservoir, and we wish to dis 
charge the same at the end of a horizontal pipe 150 feet long, and 
of the same diameter as the orifice, it will require ten times the 
head of water to accomplish it; or it maybe done by the same 
head of water, if we use a pipe about 8 inches, instead of 5 inches, 
in diameter. 

An aqueduct should be as straight as possible, not only to 
avoid unnecessary increase of length, but because the force of the 
stream is diminished by all changes of direction. If there must be 
change, it should be a gradual curve, and not an abrupt turn. 
When a pipe changes its direction by an angle, instead of a curve, 
there is a useless expenditure of force ; a change of 90 requires 
that the head of water should be increased by nearly the height 
due to the velocity of discharge. For instance, if the discharge is 
eight feet per second (which is the velocity due to one foot of fall), 
then a right angle in the pipe requires that the head of water 
should be increased by nearly one foot, in order to maintain that 

236. Jets. Since a body, when projected upward with a cer 
tain velocity, will rise to the same height as that from which it 
must have fallen to acquire that velocity, therefore, if water issue 
from the side of a vessel through a pipe bent upward, it would, 
were it not for the resistance of the air and friction at the orifice, 
rise to the level of the water in the reservoir. If water is dis 
charged from an orifice in any other than a vertical direction, it 
describes a parabola, since each particle may be regarded as a pro 
jectile (Art. 44). 

If a semicircle be described on the perpendicular side of a 
vessel as a diameter, and water 
issue horizontally from any point, 
its range, measured on the level of 
the base, equals twice the ordinate 
of that point. For, the velocity 
with which the fluid issues from 
the vessel, being that which is due 
to the height B G (Fig. 162), is 
V2j . B G (Art. 28). But after 
leaving the orifice, it arrives at the 

horizontal plane in the time in which a body would fall 

2 G D 


through G D, which is 

Since the horizontal motion 

is uniform, the space equals the product of the time by the veloc 
ity; that is, D E - \ - x V% g . B G.= 2 

G. G D = 

RIVERS. 157 

2 G If, or twice the ordinate of the semicircle at the place of dis 

The greatest range occurs when the fluid issues from the 
centre, for then the ordinate is greatest; and the range at equal 
distances above and below the centre is the same. 

The remarks already made respecting pipes apply to those 
which convey water to the jets of fire-engines and fountains. If 
the pipe or hose is very long, or narrow, or crooked, or if the jet- 
pipe is not smoothly tapered from the full diameter of the hose to 
the aperture, much force is lost by friction and other resistances, 
especially in great velocities. If the length of hose is even twenty 
times as great as its diameter, 32 per cent, of height is lost in the 
jet, and more still when the ratio of length to diameter is greater 
than this. 

237. Rivers. Friction and change of direction have great 
influence on the flow of rivers. A dynamical equilibrium, as it is 
called, exists between gravity, which causes the descent, and the 
resistances, which prevent acceleration, beyond a certain moderate 
limit ; so that, in general, the water of a river moves uniformly. 
The velocity in all parts of the same section, however, is not the 
same ; it is greatest at that part of the surface where the depth is 
greatest, and least in contact with the bed of the stream. 

To find the mean velocity through a given section, it is neces 
sary to float bodies at various places on the surface, and also below 
it, to the bottom, and to divide the sum of all the velocities thus 
obtained, by the number of observations. To obtain the quantity 
of water which flows through a given section of a river, having 
determined the velocity as above, find next the area of the section, 
by taking the depth at various points of it, and multiplying the 
mean depth by the breadth. The quantity of water is then found 
by multiplying the area by the velocity. 

The increased velocity of a stream during a freshet, while the 
stream is confined within its banks, exhibits something of the ac 
celeration which belongs to bodies descending on an inclined 
plane. It presents the case of a river flowing upon the top of 
another river, and consequently meeting with much less resistance 
than when it runs upon the rough surface of the earth itself. The 
augmented force of a stream in a freshet arises from the simulta 
neous increase of the quantity of water and the velocity. In con 
sequence of the friction of the banks and beds of rivers, and the 
numerous obstacles they meet with in their winding course, their 
velocity is usually very small, not more than three or four miles 
per hour ; whereas, were it not for these impediments, it would 
become immensely great, and its effects would be exceedingly dis- 



astrous. A very slight declivity is sufficient for giving the run 
ning motion to water. The largest rivers in the world fall about 
five or six inches in a mile. 

238. Hydraulic Pumps. The most common pumps for 
raising water operate on a principle of pneumatics, and will be de 
scribed under that subject. 

In the lifting-pump the water is pushed up in the pump tube 
by a piston placed below the water-level. In the tube A B (Fig. 
163) is a fixed valve V, a little below the 
water-level L L, while still lower is the pis 
ton P, in which there is a valve. Both of 
these valves open upward. The piston is 
attached to a rod, which extends downward 
to the frame F F. This frame can be moved 
up and down on the outside of the tube by a 
lever. "When the piston descends, the water 
passes through its valve by hydrostatic press 
ure; and when raised, it pushes the water 
before it through the fixed valve, which then 
prevents its return. In this manner, by re 
peated strokes, the water can be driven to 
any height which the instrument can bear. 

The chain pump consists of an endless 
chain with circular disks attached to it at 
intervals of a few inches, which raise the 
water before them in a tube, by means of a 
wheel over which the chain passes; the 
wheel may be turned by a crank. The 
disks cannot fit closely in the tube without causing too great re 
sistance; hence, a certain velocity is requisite in order to raise 
water to the place of discharge, ; and after the working of the 
pump ceases, the water soon descends to the level in the well. 

239. The Hydraulic Ram. When a large quantity of 
water is descending through an inclined pipe, if the lower extrem 
ity is suddenly closed, since water is nearly incompressible, the 
shock of the whole column is received in a single instant, and if 
no escape is provided, is very likely to burst the pipe. The inten 
sity of the shock of water when stopped is made the means of 
raising a portion of it above the level of the head. The instru 
ment for effecting this is called the hydraulic ram. At the lower 
end of a long pipe, P (Fig. 164), is a valve, F, opening downward ; 
near it, another valve, V, opens into the air-vessel, A-, and from 
this ascends the pipe, T, in which the water is to be raised. As 



flie valve V lies open by its weight, the water runs out, till its 
momentum at length shuts it, and the entire column is suddenly 

FIG. 1G4. 

stopped; this impulse forces the water into the air-vessel, and 
thence, by the compressed air, up the tube T. As soon as the 
momentum is expended, the valve V drops, and the process is 

240. Water- Wheels with a Horizontal Axis. The over 
shot wheel (Fig. 105) is construct 
ed with buckets on the circumfer 
ence, which receive the water just 
after passing the highest point, 
and empty themselves before 
reaching the bottom. The weight 
of the water, as it is all on one 
side of a vertical diameter, causes 
the wheel to revolve. It is usual 
ly made as large as the fall will 
allow, and will carry machinery 
with a very small supply of water, 
if the fall is only considerable. 
The moment of each bucket-full 
constantly increases from a, where 
it is filled, to F, where its acting distance is radius, and therefore 
a maximum. From F downward the moment decreases, both by 
loss of water and diminution of acting distance, and becomes zero 

The undershot wheel (Fig. 166) is re 
volved by the momentum of running water, 
which strikes the float-boards on the lower 
side. When these are placed, as in the 
figure, perpendicular to the circumference, 
the wheel may turn either way ; this is the 
construction adopted in tide-mills. When 
the wheel is required to turn only in one 
direction, an advantage is gained by placing 

FIG. 166. 



the floa -boards so as to present an acute angle toward the current? 
by which means the water acts partly by its weight, as in the over 
shot wheel. The undershot wheel is adapted to situations where 
the supply of water is always abundant. 

In the breast wheel (Fig. 167) the water is received upon the 
float-boards at about the height of the axis, and acts partly by its 
weight, and partly by its mo 
mentum. The planes of the 
float-boards are set at right 
angles to the circumference of 
the wheel, and are brought so 
near the mill-course that the 
water is held and acts by its 
weight, as in buckets. 

FIG. 167. 

FIG. 168. 

241. The Turbine. This 

very efficient water-wheel, fre 
quently called the French tur 
bine, is of modern invention, 
and has received its chief im 
provements in this country. It revolves on a vertical axis, and 
surrounds the bottom of the reservoir from which it receives the 
water. The lower part of the reservoir is divided into a large 
number of sluices by curved partitions, which direct the water 
nearly into the line of a tangent, as it issues upon the wheel. The 
vanes of the wheel are curved in the opposite direction, so as to 
receive the force of the issuing streams at right angles. The hori 
zontal section (Fig. 168) shows 
the lower part of the reservoir 
with its curved guides, a, a, a, 
and the wheel with its curved 
vanes, v, v, v, surrounding the 
reservoir ; D is the central tube, 
through which the axis of the 
wheel passes. Fig. 169 is a ver 
tical section of the turbine ; but 
it does not present the guides 
of the reservoir, nor the vanes 
of the wheel. C G, C G, is the 
outer wall of the reservoir; 
Z), D, its inner wall or tube ; 
F F, the base, curved so as to 

turn the descending water gradually into a horizontal direction. 
The outer wall, which terminates at G Gr, is connected with the 
base and tube by the guides which are shown at a, a, in Fig, 168. 



The lower rim of the wheel, //, //, is connected with the upper 
rim, P 9 P, by the vanes between them, r 9 v (Fig. 168), and to the 

FIG. 1G9. 

axis, E, Uj by the spokes /, /. The gate, /, J 9 is a thin cylinder 
which is raised or lowered between the wheel and the sluices of 
the reservoir. The bottom of the axis revolves in the socket 7f, 
and the top connects with the machinery. As the reservoir can 
not be supported from below, it is suspended by flanges on the 
masonry of the wheel-pit, or on pillars outside of the wheel. 
To prevent confusion in the figure, the supports of the reservoir 
and the machinery for raising the gate are omitted. By the 
curved base and guides of the reservoir, the water is conducted in 
a spiral course to the wheel, with no sudden change of direction, 
and thus loses very little of its force. The wheel usually runs 
below the level of the water in the wheel-pit, as represented in 
the figure, L L being the surface of the water. The reservoir is 
sometimes merely the extremity of a large tapering tube or supply 
pipe, bent from a horizontal to a vertical direction. In such a 
case, the tube D D, in which the axis runs, passes through the 
upper side of the supply pipe. The figure represents only the 
lower part. 

242. Barker s Mill. This machine operates on the principle 

of unbalanced hydrostatic pressure. It consists of a vertical hollow 

cylinder, A B (Fig. 170), free to revolve on its axis M N 9 and 

having a horizontal tube connected with it at the bottom. Near 





each end of the horizontal tube, at P and P , is an orifice, one on 
one side, and one on the opposite. The FIG 170. 

cylinder, being kept full of water, whirls 
in a direction opposite to that of the dis 
charging streams from P and P . This 
is owing to the fact that hydrostatic 
pressure is removed from the apertures, 
while on the interior of the tube, at 
points exactly opposite to them, are 
pressures which are now unbalanced, 
but which would be counteracted by 
the pressures at the apertures, if they 
were closed. The centrifugal force, 
after the machine is in rotation, has 
the effect to increase the pressure, and 
therefore the speed of rotation. 

243. Resistance to Motion in a Liquid. The resistance 
which a body encounters in moving through any fluid arises from 
the inertia of the particles of the fluid, their want of perfect mo 
bility among each other, and friction. Only the first of these 
admits of theoretical determination. So far as the inertia of the 
fluid is concerned, the resistance which a surface meets with in 
moving perpendicularly through it varies as the square of the ve 
locity. For the resistance is measured by the momentum imparted 
by the moving body to the fluid. And this momentum (m) varies 
as the product of the quantity of fluid set in motion (q), and its 
velocity (v) ; or m cc q v. But it is obvious that the quantity dis 
placed varies as the velocity of the body, or q x v\ hence m oc v 1 . 
Therefore the resistance varies as the square of the velocity. 

This proposition is found to hold good in practice, where the 
velocity is small, as the motions of boats or ships in water ; but 
vrhen the velocity becomes very great, as that of a cannon ball, 
the resistance increases in a much higher ratio than as the square 
of the velocity. Since action and reaction are equal, it makes no 
difference, in the foregoing proposition, whether we consider the 
body in motion and the fluid at rest, or the fluid in motion and 
striking against the body at rest. 

Since the resistance increases so rapidly, there is a wasteful 
expenditure of force in trying to attain great velocities in naviga 
tion. For, in order to double the velocity of a steamboat, the 
force of the steam must be increased four fold ; and in order to 
triple its velocity, the force must become nine times as great. 

When the resistance becomes equal to the moving force, the 
body moves uniformly, and is said to be in a state of dynamical 



equilibrium. Thus, a body falling freely through the air by 
gravity does not continue to be accelerated beyond a certain 
limit, but is finally brought, by the resistance of the air, to a uni 
form motion. 

244. Water Waves. These are moving elevations of water, 
caused by some force which acts unequally on its surface. There 
are two very different kinds of waves, called, respectively, wares 
of oscillation and waves of translation. In the first kind the par 
ticles of water have a vibratory or reciprocating motion, by which 
the vertical columns are alternately lengthened and shortened. A 
familiar example of this kind is the sea-wave. In the waves of 
translation the particles are raised, transferred forward, and then 
deposited in a new place, without any vibratory movement. 

245. Waves of Oscillation. If a pebble be tossed upon 
still water, it crowds aside the particles beneath it, and raises them 
above the level, forming a wave around it in the shape of a ring. 
As soon as this ring begins to descend, it elevates above the level 
another portion around itself, and thus the ring-wave continues 
to spread outward every way from the centre. But in the mean 
time the water at the centre, as it rises toward the level, acquires 
a momentum which lifts it above that level. From that position it 
descends, and once more passes below the level, thus starting a 
new wave around it, as at first, only of less height. Hence, we see 
as the result of the first disturbance, a series of concentric waves 
continually spreading outward and di 
minishing in height at greater dis 
tances, until they cease to be visible. 

In Fig. 171 are represented three circu 
lar waves at one of the moments of 
time when the centre is lowest. The 
shaded parts are the basins or troughs, 
and the light parts, c, c, c, are the ridges 
or crests. Fig. 172 is a vertical section 
along the line c, c, through the centre 
of the system, corresponding to the mo 
mentary arrangement of Fig. 171. The central basin is at b, and 
the crests at c, c, c. A little later, when either crest has moved 
half way to the place of the next one, 
both figures will have become reversed ; 
the centre will be a hillock, the troughs 
will be at c, c, and the crests at the 
middle points between them. 

Except in the circular arrangement of the crests and troughs 
around a centre, the waves of the foregoing experiment illustrate 

FIG. 171. 

FIG. 172. 

c c 


common sea- waves. They constitute a system of elevations and 
depressions moving along the surface at right angles to the line 
of the wave-crest. 

246. Phases. In the cross-section (Fig. 172), where the 
waves are shown in profile, any particular part of the curve is 
called a phase. Different phases are generally unlike, both in ele 
vation and in movement. The corresponding parts of different 
waves are called like phases ; and those points in which the mo 
lecular motions are reversed are called opposite phases. The 
highest, points of the crests of two waves are like phases; the 
highest point of the crest and the lowest point of the trough are 
opposite phases. Two points half way from crest to trough, one 
on the front of the wave, and the other on the rear of it, are also 
opposite phases, although they are at the same elevation ; for they 
are moving in opposite directions. The length of a wave is the 
horizontal distance between two successive like phases. 

247. Molecular Movements. The water which constitutes 
a system of waves does not advance along the surface, as the waves 
themselves do ; for a floating body is not borne along by them, but 
alternately rises and falls as the waves pass under it. Each par 
ticle of water, instead of advancing with the wave, oscillates about 
its mean place, alternately rising as high as the crest, and falling 
as low as the trough. Its path is the circumference of a vertical 
circle. Let B B (Fig. 173) represent two successive troughs, and 

Fro. 173. 


a b c d e 

O the intervening crest; and for convenience suppose a a , the 
wave length, to be divided into eight equal parts. The waves 
move in the direction of the straight arrow, while the particles of 
water revolve in the direction of the bent arrows. The points 
1, 2, 3, &c., represent particles which, if the water were at rest, 
would be directly above the points a, b, c, &c. At the moment 
represented, 1 is at the extreme left of its revolution, 2 is at 45 
below, 3 at the lowest point, &c. When the wave has advanced 
one-eighth of its length, 1 will have ascended 45, 2 will have as 
cended to the extreme left, and each of the eight particles will 


have revolved one-eighth of the circumference shown in the figure. 
Then 4 will be at the bottom, and 8 at the top. Each particle of 
water on the front of the wave, from 1 to 3, and from 7 to 3 , is 
ascending ; each one on the rear, from 3 to 7, is descending. It 
is plain that while the wave advances its whole length, that is, 
while the phase B is moving to B , each particle makes a complete 
revolution; 3 , which is now lowest, will be lowest again, having 
in the meantime occupied all other points of the circumference. 

Particles below the surface, as far as the wave disturbance 
reaches, perform synchronous revolutions, but in smaller circles, 
as represented in the figure. 

248. Form of Waves of Oscillation. The sectional form 
of these waves is that of the inverted trochoid, a curve described 
by a point in a circle as it rolls on a straight line. The curvature 
of the crest is always greater than that of the trough, and the 
summit may possibly be a sharp ridge, in which case the section 
of the trough is a cycloid, the describing point of the rolling circle 
being on the circumference; the height of such waves is to their 
length as the diameter of a circle to the circumference. If waves 
are ever higher than about one-third of their length, the summits 
are broken into spray. 

249. Distortion of the Vertical Columns. Where the 
surface is depressed below its level, some of the water must be 
crowded laterally out of its place, and the vertical columns, being 
shorter, must necessarily be wider, at least in the upper part. So, 
too, where the surface is raised above its level, the lengthened 
columns must be narrower. In Fig. 173 these effects are made 
apparent as the necessary result of the revolutions of the particles. 
The dotted lines, 1 , 2 b, 3 c, &c., were all vertical lines when the 
water was at rest. But now they are swayed, some to the right 
and some to the left, none being vertical, except under the highest 
and lowest points of the waves. Under the trough the lines are 
spread apart, and under the crest they are drawn together. The 
sectional figures 1 a b 2, 2 I c 3, &c., which would all be rectangu 
lar if the water were at rest, are now distorted in form, the upper 
parts being alternately expanded and contracted in breadth as the 
successive phases pass them. 

250. Sea-Waves. The waves raised by the wind rarely ex 
hibit the precise forms above described, and the particles rarely 
revolve in exact circles, partly because there is scarcely ever a sys 
tem of waves undisturbed by other systems, which are passing 
over the water at the same time, and partly because the wind, 
which was the original cause of the waves, acts continually upon 
their surfaces to distort and confuse them. 


The interference of waves denotes, in general, the resultant 
system, which is produced by the combination of two or more 
separate systems. The joint effect of two systems is various, ac 
cording as they are more or less unlike as to length of waves. 
But even if two systems are just alike, still the effect of interfer 
ence will vary, according to the coincidence or the degree of dis 
crepancy of their like phases. For instance, if two similar 
systems exactly coincide, phase for phase, the waves simply have 
double height ; or, in general terms, there is double intensity in 
the wave motion. But if the phases of one system exactly coincide 
with the opposite phases of the other, then the water is nearly 
level, the crests of each system filling the troughs of the other. 
These two effects may be plainly seen in the intersections of ring- 
waves formed by dropping two pebbles on still water. 

251. Waves of Translation. The principal characteristics 
of the wave of translation are, that it is solitary i. e., it does not 
belong to a system, like the other kind ; and that its length and 
velocity both depend on the depth of the water. Where the water 
is deeper, the wave travels faster, and its length (measured in the 
direction of its progress) is longer. A wave of this character is 
stc^rted in a canal by a moving boat; and when the boat stops, it 
moves on alone. A grand example of this species is found in the 
tide-wave of the ocean. It is called the wave of translation be 
cause the particles of water are borne forward a certain distance 
while the wave is passing, and then remain at rest. 




252. Gases Distinguished from Liquids. The property 
of mobility of particles, which belongs to all fluids, is more re 
markable in gases than in liquids. 

While gaseous substances are compressed with ease, they are 
always ready to expand and occupy more space. This property, 
called dilatability, scarcely belongs to liquids at all. The force 
which gases show in expanding is called tension. 

253. Change of Condition. Liquids, and even solids, may 
be changed into the gaseous or aeriform condition by heating 
them sufficiently. By being cooled, they return again to their 
former state. In the gaseous form they are called vapors. And 
some substances which are ordinarily gases can be so far cooled, 
especially under great pressure, as to be reduced to the liquid or 
solid form. Those which have never been thus reduced are called 
permanent gases. 

The mechanical properties of the gases may all be illustrated 
by experiments performed upon atmospheric air. 

254. Mariotte s Law. 

At a given temperature, the volume of air is inversely as the 
compressing force. 

An instrument constructed for showing this is called Mariotte s 
tube. The end B (Fig. 174) is sealed, and A open. Pour in small 
quantities of mercury, inclining the tube so as to let air in or out, 
till both branches are filled to the zero point The air in the 
short branch now has the same tension as the external air, since 
they just balance each other. If mercury be poured in till tin 
column in the short tube rises to 6 , the inclosed air is reduced to 
one-half of its original volume, and the column A in the long branch 
is found to be 29 or 30 inches above the level of C, according to 



FIG. 174 

the barometer at the time. Thus, two atmospheres, one of mer 
cury, the other of air above it, have compressed the inclosed air 
into one-half its volume* If the tube is of 
sufficient length, let mercury be poured in 
again, till the air is compressed to one-third 
of its original space; the long column, 
measured from the level of the mercury in 
the short one, is now twice as high as be 
fore ; that is, three atmospheres, two of mer 
cury and one of air, have reduced the same 
quantity of air to one-third of its first vol 
ume. This law has been found to hold good 
in regard to atmospheric air up to a pressure 
of nearly thirty atmospheres. 

On the other hand, if the pressure on a 
given mass of air is diminished, its volume 
is found to increase according to the same 
law. When the pressure is half an atmo 
sphere, the volume is doubled ; when one- 
third of an atmosphere, the volume is three 
times as great, &c. 

This law is found, however, not to be 
strictly applicable to all the gases. Some 
are compressed a little more, and others a 
little less, than Marietta s law would re 

Since the tension of the inclosed air 
always ^balances the compressing force, and 
since the density is inversely as the volume, 
it follows from Mariotte s law that when the temperature is the same, 

Tlie tension of air varies as the compressing f orce ; and 

TJie tension of air varies as its density. 

255. The Air-Fuinp, This is an instrument by which 
nearly all the air can be removed from a vessel or receiver. It 
has a variety of forms, one of which is shown in Pig. 175. In the 
barrel B an air-tight piston is alternately raised and depressed by 
the lever, the piston-rod being kept vertical by means of a guide. 
The pipe P connects the bottom of the barrel with the brass plate 
//, on which rests the receiver R. The surface of the plate and 
the edge of the receiver are both ground to a plane. G is the 
gauge which indicates the degree of exhaustion. There are three 
valves, the first at the bottom of the barrel, the second in the 
piston, and the third at the top of the barrel. These all open up 
ward, allowing the air to pass out, but preventing its return. 


FIG. 176. 

256. Operation. When the piston is depressed, the air below 
it, by its increased tension, presses down the first valve, and opens 
the second, and escapes into the upper part of the barrel. When 
the piston is raised, the air above it cannot return, but is pressed 
through the third valve into the open air ; while the air in the re 
ceiver and pipe, by its tension, opens the 
first valve, and diffuses itself equally 
through the receiver and barrel. An 
other descent and ascent only repeat 
the same process; and thus, by a suc 
cession of strokes, the air is nearly all 

The exhaustion can be made more 
complete if the first and second valves 
are opened by the action of the piston 
and rod, rather than by the tension of 
the air. This method is illustrated by 
Fig. 176, a section of the barrel and pis 
ton. The first and second valves, as 
shown in tKe figure, are conical or pup 
pet valves, fitting into conical sockets. 
The first has a long stem attached, which 


passes through the piston air-tight, and is pulled up by it a little 
way, till it is arrested by striking the top of the barrel. The sec 
ond valve is a conical frustum on the end of the piston-rod. When 
the rod is raised, it shuts the valve before moving the piston ; 
when it begins to descend, it opens the valve again before giving 
motion to the piston. The first valve is shut by a lever, which 
the piston strikes at the moment of its reaching the top. The oil 
which is likely to be pressed through the third valve is drained off 
by the pipe (on the right in both figures) into a cup below the 

257. Rate of Exhaustion. The quantity removed, by suc 
cessive strokes, and also the quantity remaining in the receiver, 
diminishes in the same geometrical ratio. For, of the air occupy 
ing the barrel and receiver, a barrel-full is removed at each stroke, 
and a receiver-full is left. If, for example, the receiver is three 
times as large as the barrel, the air occupies four parts before the 
descent of the piston; and by the first stroke one-fourth is re 
moved, and three-fourths are left. By the next stroke, three- 
fourths as much will be removed as before (| of |, instead of J of 
the whole), and so on continually. The quantity left obviously 
diminishes also in the same ratio of three-fourths. In general, if 
b expresses the capacity of the barrel, and r that of the receiver 

and connecting-pipe, the ratio of each descending series is T - 

With a given barrel, the rate of exhaustion is obviously more 
rapid as the receiver is smaller. If the two were equal, ten strokes 
would rarefy the air more than a thousand times. For (|) 10 = 


As a term of this series can never reach zero, a complete ex 
haustion can never be effected by the air-pump ; but in the best 
condition of a well-made pump, it is not easy to discover by the 
gauge that the vacuum is not perfect. 

258. Experiments with the Air-Pump. By the air-pump 
a great variety of experiments may be performed, illustrative of 
the mechanical properties of the air. The obstruction of the air 
being removed, light and heavy bodies are seen to fall with equal 
rapidity ; a wheel with vanes perpendicular to the plane of rota 
tion runs as freely as if they coincided with that plane, and water 
boils below blood-heat. The iveiglit of a given volume of air is 
obtained by first weighing a vessel filled with air, and then empty. 
The pressure of air in every direction is rendered apparent by 
many striking effects, such as lifting weights, holding together the 
Magdeburg hemispheres, and throwing jets of water ; also, by the 
difference of pressures on the upper and lower side, bodies are 



FIG. 177. 

shown to weigh less in air than in a vacuum. And, finally, the 
tension or expansive force is exhibited by experiments equally nu 
merous and interesting. 

259. The Air Condenser. While the air-pump shows 
the tendency of air to dilate indefinitely, as the com 
pressing force is removed, another useful instrument, 
the condenser, exhibits the indefinite compressibility 
of air. Like the pump, it consists of a barrel and piston, 
but its valves, one in the piston and one at the bottom 
of the barrel, open downward. Fig. 177 shows the exterior 
of the instrument. If it be screwed upon the top of a 
strong receiver (Fig. 178), with a stop-cock connecting 
them, air may be forced in, and then secured by shutting 
the stop-cock. When the piston is depressed, its own 
valve is shut by the increased tension of the air beneath 
it, and the lower one opened by the same force. When 
the piston is raised, the lower valve is kept shut by the 
condensed air in the receiver, and that of the piston is 
opened by the weight of the outer air, which thus gets 
admission below the piston. 

The quantity of air in the receiver increases at each 
stroke in an arithmetical ratio, because the same quan 
tity, a barrel r full of common air, is added every time the piston 
is depressed. A small Marietta s tube is attached to the receiver, 
to show how many atmospheres have been ad 
mitted. Fl - 178. 

260. Experiments with the Air Con 
denser. If the receiver be partly filled with 
water, and a pipe from the stop-cock extend into 
it, then when the condenser has been used and 
removed, and the stop-cock opened, a jet of water 
will be thrown to a height corresponding to the 
tension of the inclosed air. A gas-bag being 
placed in the condenser, then filled and shut, will 

become flaccid when the air around it is compressed. A thin glass 
bottle, sealed, will be crushed by the same force. By these and 
other experiments may be shown the effects of increased tension. 

261. Torricelli s Experiment A glass tube A B (Fig. 179) 
about three feet long, and hermetically sealed at one end, is filled 
with mercury, and then, while the finger is held tightly on the 
open end, it is inverted in a cup of mercury. On removing the 
finger after the end of the tube is beneath the surface of the mer 
cury, the column sinks a little way from the top, and there re 
mains. Its height is found to be nearly thirty inches above the 



level of mercury in the cup. If sufficient care is taken to expel 
globules of air from the liquid, the 
space above the column in the tube is FlG - 

as perfect a vacuum as can be obtained. 
It is called the Torricellian vacuum, 
from Torricelli of Italy, a disciple of 
Galileo, who, by this experiment, dis 
proved the doctrine that nature abhors 
a vacuum, and fixed the limits of at 
mospheric pressure. 

232. Pressure of Air Meas 
ured. The column is sustained in 
the Torricellian tube by the pressure 
of air on the surface of mercury in the 
vessel ; for the level of a fluid surface 
cannot be preserved unless there is an 
equal pressure on every part. Hence, 
the column of mercury on one part, 
and the column of air on every other 
equal part, must press equally. To de 
termine, therefore, the pressure of air, 
we have only to weigh the column of 
mercury, and measure the area of the 

mouth of the tube. If this is carefully done, it is found that the 
weight of mercury is about 14.7 Ibs. on a square inch. Therefore 
the atmosphere presses on the earth with a force of nearly 15 
pounds to every square inch, or more than 2000 Ibs. per square 

The specific gravity of mercury is about 13.6; and therefore 
the height of a column of water in a Torricellian tube should be 
13.6 times greater than that of mercury, that is, about 34 feet. 
Experiment shows this to be true. And it was this significant 
fact, that equal weights of water and mercury are sustained in 
these circumstances, which led Torricelli to attribute the effect to 
a common force, namely, the pressure of the air. 

263. Pascal s Experiment. As soon as Torricelli s discov 
ery was known, Pascal of France proposed to test the correct 
ness of his conclusion, by carrying the apparatus to the top of a 
mountain, in order to see if less air above the instrument sustained 
the mercury at a less height. This was found to be true; the 
column gradually fell, as greater heights were attained. The ex 
periment of Pascal also determined the relative density of mercury 
and air. For the mercury falls one-tenth of an inch in ascending 
87.2 feet; therefore the weight of the one-tenth of an inch of mer- 


cury was balanced by the weight of the 87.2 feet of air. There 
fore the specific gravities of mercury and air (being inversely as 
the heights of columns in equilibrium) are as (87.2 x 12 x 10 ) 
10464 : 1. In the same way it is ascertained that water is 770 
times as dense as air. These results can of course be confirmed 
by directly weighing the several fluids, which could not be done 
before the invention of the air-pump. 

264. The Barometer. When the Torricellian tube and 
basin are mounted in a case, and furnished with a graduated scale, 
the instrument is called a barometer. The scale is divided into 
inches and tenths, and usually extends from 26 to 32 inches, a 
space more than sufficient to include all the natural variations in 
the weight of the atmosphere. By attaching a vernier to the 
scale, the reading may be carried to hundredths and thousandths 
of an inch, as is commonly done in meteorological observations. 
By observing the barometer from day to day, and from hour to 
hour, it is found that the atmospheric pressure is constantly fluc 

As the meteorological changes of the barometer are all com 
prehended within a range of two or three inches, much labor has 
been expended in devising methods for magnifying the motions 
of the mercurial column, so that more delicate changes of atmo 
spheric pressure might be noted. The inclined tube and the wheel 
barometer are intended for this purpose. A description of these 
contrivances, however, is unnecessary, as they are all found to be 
inferior in accuracy to the simple tube and basin. 

265. Corrections for the Barometer. 

1. For change of level in the basin. The numbers on the 
barometer scale are measured from a certain zero point, which is 
assumed to be the level of the mercury in the basin. If now the 
column falls, it raises the surface in the basin ; and if it rises, it 
lowers it. If the basin is broad, the change of level is small, but 
it always requires a correction. To avoid this source of error, the 
bottom of the basin is made of flexible leather, with a screw under 
neath it, by which the mercury may be raised or lowered, till its 
surface touches an index that marks the zero point. This adjust 
ment should always be made before reading the barometer. 

2. For capillarity. In a glass tube mercury is depressed by 
capillary action (Art. 227). The amount of depression is less as 
the tube is larger. This error is to be corrected by the manufac 
turer, the scale being put below the true height by a quantity 
equal to the depression. 

There is a slight variation in this capillary error, arising from 


the fact that the rounded summit of the column, called the menis 
cus, is more convex when ascending than when descending. To 
render the meniscus constant in its form, the barometer should be 
jarred before each reading. 

3. For temperature. As mercury is expanded by heat and con 
tracted by cold, a given atmospheric pressure will raise the column 
too high, or not high enough, according to the temperature of the 
mercury. A thermometer is therefore attached to the barometer, 
to show the temperature of the instrument. By a table of correc 
tions, each reading is reduced to the height the mercury would 
have if its temperature was 32 F. 

4. For altitude of station. Before comparing the observations 
of different places, a correction must be made for altitude of sta 
tion, because the column is shorter according as the place is 
higher above the sea level. 

266. The Aneroid Barometer. This is a small and port 
able instrument, in appearance a little like a large chronometer. 
The essential part of this barometer is a flat cylindrical metallic 
box, whose upper surface is corrugated, so as to be yielding. The 
box being partly exhausted of air, the external pressure causes the 
top to sink in to a certain extent ; if the pressure increases, the 
surface descends a little more; if it diminishes, a little less. These 
small movements are communicated by a system of levers to an 
index on the graduated face of the barometer. The box and levers 
are concealed and protected within the outer case. As might be 
expected, its range is limited, and its indications not perfectly re 
liable ; but for obtaining results in which accuracy is not essen 
tial, its lightness and convenient form and size recommend it, 
especially for portable uses. 

267. Pressure and Latitude. The mean pressure of the 
atmosphere at the level of the sea is very nearly 30 inches. But it 
is not the same at all latitudes. From the equator either north 
ward or southward, the mean pressure increases to about latitude 
30, by a small fraction of an inch, and thence decreases to about 
65, where the pressure is less than at the equator, and beyond 
that it slightly increases. This distribution of pressures in zones 
is due to the great atmospheric currents, caused by heat in con 
nection with the earth s rotation on its axis. 

The amount of variation in barometric pressure is very unequal 
in different latitudes ; and in general, the higher the latitude, the 
greater the variation. Within the tropics the extreme range 
scarcely ever exceeds one-fourth of an inch, while at latitude 40 
it is more than two inches, and in higher latitudes even reaches 
three inches. 


268. Diurnal Variation. If a long series of barometric ob 
servations be made, and the mean obtained for each hour of tin 
day, the changes caused by weather become eliminated, and the 
diurnal oscillation reveals itself. It is found that the pressure 
reaches a maximum and a minimum twice in 24 hours. The 
times of greatest pressure are from 9 to 10, and of least pressure 
from 3 to 4, both A. M. and P. M. In tropical climates this varia 
tion is very regular, though small ; but in the temperate zones the 
irregular fluctuations of weather conceal it in a great degree. 

This double oscillation is the mingled effect of heat and 
moisture, each of which alone would produce a single oscillation 
extending through the entire day. 

In some countries of the torrid zone there is a regular annual 
oscillation of the barometer; but in the temperate zones this is 
scarcely perceptible. 

269. The Barometer and the Weather. The changes in 
the height of the barometer column depend directly on nothing 
else than the atmospheric pressure. But these changes of pressure 
are due to several causes, such as wind and changes of temperature 
and moisture. 

The practice formerly prevailed of engraving at, different points 
of the barometer scale several words expressive of states of weather, 
"fair, rain, frost, wind," &c. But such indications are worthless, 
being as often false as true ; this is evident from the fact that the 
height of the column would be changed from one kind of weather 
to another by simply carrying the instrument to a higher or lower 

No general system of rules can be given for anticipating 
changes of weather by the barometer, which would be applicable 
in different countries. Eules found in English books arc of very 
little value in America. 

Severe and extensive storms are almost always accompanied by 
a fall of the barometer while passing, and succeeded by a rise of 
the barometer. 

270. Heights Measured by the Barometer. Since mer 
cury is 10464 times as heavy as air (Art. 2G3), if the barometer is 
carried up until the mercury falls one inch, it might be inferred 
that the ascent is 104G4 inches, or 87 feet. This would be the 
case if the density were the same at all altitudes. But, on account 
of diminished pressure, the air is more and more expanded at 
greater heights. Besides this, the height due to a given fall of the 
mercury varies for many reasons, such as the temperature of the 
air, the temperature of the mercury, the elevation of the stations, 
and their latitude. Hence, the measurement of heights by the 


barometer is somewhat troublesome, and not always to be relied 
on. Formulae and tables for this purpose are to be found in prac 
tical works on physics. 

271. The Gauge of the Air-Pnmp. The Torricellian tube 
is employed in different ways as a gauge for the air-pump, to indi 
cate the degree of exhaustion. In Fig. 175 the gauge G is a tube 
about 33 inches long, both ends of which are open, the lower im 
mersed in a cup of mercury, and the upper communicating with 
the interior of the receiver. As the exhaustion proceeds, the 
pressure is diminished within the tube, and the external air raises 
the mercury in it. A perfect vacuum would be indicated by a 
height of mercury equal to that of the barometer at the time. 

Another kind of gauge is a barometer already filled, the basin 
of which is open to the receiver. As the tension of air in the re 
ceiver is diminished, the column descends, and would stand at the 
same level in both tube and basin, if the vacuum were perfect. 

A modified form of the last, called the siphon gauge, is the 
best for measuring the rarity of the air in the receiver when the 
vacuum is nearly perfect. Its construction is shown by Fig. 180. 
The top of the column, A, is only 5 or 6 inches above 
the level of B in the other branch of the recurved tube. 
As the air is withdrawn from the open end C 9 the ten 
sion at length becomes too feeble to sustain the col 
umn ; it then begins to descend, and the mercury in 
the two branches approaches a common level. 

272. Buoyant Power of the Air. If a large 
and a small body are in equilibrium on the two arms 
of a balance, and the whole be set under a receiver, 
and the air be removed, the larger body will prepon 
derate, showing that .it is really the heaviest. Their apparent 
equality of weight when in the air is owing to its buoyant power ; 
for air, like water and all fluid substances, diminishes the apparent 
weight of an immersed body by just the weight of the displaced 
fluid. Hence, the larger the body, the more weight it loses. 

It follows that if a body weighs less than the displaced air, it 
will rise just as light bodies do in water. It is in this way that 
balloons are made to ascend. By the use of a large volume of hy 
drogen, inclosed in a silk envelope, rendered air-tight by varnish, 
a car with several persons in it can be carried to a great height. 
The greatest height ever attained is about 23000 feet, or nearly 
4.5 miles. The mercury of a barometer at that height falls to 12.5 





273. Pneumatic Instruments. Besides the apparatus de 
scribed in the foregoing chapter, by the aid of which the proper 
ties of the air are discovered, there are several articles in common 
use whose utility depends more or less on the same properties, and 
which serve as good illustrations of the principles already pre 

274. The Bellows. The simple or Jiand-bellows consists of 
two boards or lids hinged together, and having a flexible leather 
round the edges, and a tapering tube through which the air is 
driven out. In the lower board there is a hole with a valve lying 
on it, which can open inward. On separating the lids, the air by 
its pressure instantly lifts the valve and fills the space between 
them; but when they are pressed together, the valve shuts, and 
the air is compelled to escape through the pipe. The stream is 
intermittent, passing out only when pressure is applied. 

The compound bellows, used for forges where a constant stream 
is needed, are made with two compartments. The partition G T 
(Fig. 181) is fixed, and 

has in it a valve V FIG. 181. 

opening upward. The 
lower lid has also a 
valve V opening up 
ward, and the upper 
one is loaded with 
weights. The pipe T 
is connected with the 
upper compartment. 
As the lower lid is 
raised by the rod A JB, 
which is worked by the 
lever E B, the air in 
the lower part is crowded through V into the upper part, whence 
it is by the weights pressed through the pipe I 7 in a constant 
stream. When the lower lid falls, the air enters the lower com 
partment by the valve V. 

275. The Siphon. If a bent tube A B G (Fig. 182) be filled, 
and one end immersed in a vessel of water, the liquid will be dis- 




FIG. 182. 

charged through the tube so long as the outer end is lower than 
the level in the vessel. Such a tube is called a siphon, and is 
much used for removing a liquid from the top of a reservoir with 
out disturbing the lower part. The height 
of the bend B above the fluid level must 
be less than 34 feet for water, and less 
than 30 inches for mercury. The reasons 
for the motion of the water are, that the 
atmosphere is able to sustain a column 
higher than E B, and that C B is longer 
than E B. The two pressures on the 
highest cross-section B of the tube are 
unequal. For the atmospheric pressure 
at E is able to sustain 34 feet of water, 
and therefore at B exerts a pressure equal 
to 34 - E B toward the right. At C the 
air also presses upward with a force equal 
to 34 feet, and therefore at B it exerts a 
pressure to the left equal to 34 C B. 
Subtracting 34 C B from 34 E B } 
we have the remainder, C D, for the ex 
cess of pressure to the right, or outward 
from the vessel. Therefore the water will 
flow with a velocity due to the weight of 
D C\ hence, the velocity diminishes as 
the vessel empties. 

If the tube is small, it may be filled by suction, after the end A 
is immersed. If it is large, it may be inverted and filled, and then 
secured by stop-cocks, till the end is beneath the water. 

276. Intermitting Springs. Springs which flow freely for a 
time, and then cease for a certain interval, after which they flow 
again, are found in some cases to operate on the principle of the 
siphon. Suppose a reservoir or hollow in the interior of a hill, 
having a siphon-shaped outlet. It is obvious, upon hydrostatic 
principles, that no water will be discharged until the fluid has 
reached a level in the reservoir as high as the top of the bend in 
the outlet. Then it will begin to run out, and will continue to 
run until the water has descended to the level of the outlet ; after 
which no more water will be discharged until enough has collected 
to reach the higher level, as before. 

277. The Suction Pump. The section (Fig. 183) exhibits 
the construction of the common suction pump. By means of a 
lever, the piston P is moved up and down in the tube A V. In 



FIG. 183. 

the piston is a valve opening upward, and at the top of the pipe 

H G is another valve, also opening upward. The latter must be 

at a less height than 34 feet above the water C. When the piston 

is raised, its valve is kept shut by the weight 

of air ahove, and the atmospheric pressure at 

C lifts a column of water C H to such a height 

that its weight, added to the tension of the 

rarefied air, HP, equals 34 feet of water. 

When P descends, the air below is prevented 

from returning by the lower valve, and escapes 

through the piston. The piston being raised 

again, the water rises still higher, till at length 

it passes through the valve, and the piston dips 

into it; after this it is lifted directly to the 

discharge pipe 8, without the intervention of 

the air. 

278. Calculation of the Force. Let 

the whole atmospheric pressure be represented 
by 34 (its equivalent in feet of water), and the 
height of water, C H, by li. Since the tension 
of air in the tube, added to 7i, equals 34, there 
fore the tension = 34 li\ and this force is 
exerted upward on the lower side of the piston ; 
while the downward pressure on the top of the 
piston 34. The difference of the two = 7i, 
which is, therefore, the height of water, whose 
downward pressure is to be overcome. We ar 
rive at the same result if the water is above "_n"Z_~l_"i-"Z-"r 
the piston at A, when h = A C. For, in this 
case, the pressure upward on P 34 P C; while the down 
ward pressure = 34 4- A P ; and the difference between them is 
P C + A P - h. Therefore, in every case, the force required to 
lift the piston and column of water is that which would be re 
quired to lift the same weight in any other way. The atmosphere 
has no other agency than to furnish a convenient mode of apply 
ing the force. 

If d = the diameter of the piston, in decimals of a foot, then 
I rr d* = its area; j n d* h = the cubic feet of water; and 
} TT d* li x 62.5 the pounds of water. 


279. The Forcing Pump. The piston of the forcing pump 
(Fig. 184) is solid, and the upper valve V opens into the side 
pipe V S. In the ascent of the piston, the water is raised as in 
the suction pump ; but in its descent, a force must be applied 



FIG. 184. 

to press the water which is above V into the side pipe 
through V. 

As in the suction pump, the force expend 
ed is that required to lift \ TT d? li x 62.5 
pounds of water. But the two differ in this 
respect : in the suction pump the force is all 
expended in raising the piston ; in the forcing 
pump the force is divided, and the column 
below P is lifted while the piston ascends, 
and that above P while it descends. 

The piston is only one of many contriv 
ances for producing rarefaction of air in a 
pump-tube; but since it is the most simple 
and most easily kept in repair, the piston- 
pump is generally preferred to any other. 

280. The Fire-Engine. This machine 
generally consists of one or more forcing 
pumps, with a regulating air-vessel, though 
the arrangement of parts is exceedingly varied. 
Fig. 185 will illustrate the principles of its 
construction. As the piston, P, ascends, the 
water is raised through the valve, V, by at 
mospheric pressure. As P descends, the water 
is driven through F into the air-vessel, M, 
whence by the condensed air it is forced out 
without interruption through the hose-pipe, L. 
The piston P f operates in the same way by 

alternate movements. The piston-rods are attached to a lever 
(not represented), to which the strength of several men can be ap 
plied at once by means of hand-bars called brakes. 

The air-vessel may be attached 
to any kind of pump, whenever it 
is desired to render the stream con 

281. Hero s Fountain. The 

condensation in the air-vessel, from 
which water is discharged, may be 
produced by the weight of a column 
of water. An illustration is seen in 
Hero s fountain, Fig. 186. A ver 
tical column of water from the ves 
sel, A, presses into the air-vessel, B, 
and condenses the air more or less, 
accprding to the height of A B. 

FIG. 185. 



From the top of this vessel an air-tube conveys the force of the 
compressed air to a second air-vessel, C, 
which is nearly full of water, and has a jet- 
pipe rising from it Since the tension of 
air in C is equal to that in B, a jet will be 
raised which, if unobstructed, would be 
equal in height to the compressing column, 

This plan has been employed to raise 
water from a mine in Hungary, and hence 
called " the Hungarian machine." 

282. Manometers. These are instru 
ments for measuring the tension of gases 
or vapors. In one kind of manometer the 
law of Mariotte is employed. The tube 
A B (Fig. 187), closed at the top, has its 
open end beneath the surface of mercury 
in the closed cistern C. The vessel D, 
containing the gas or vapor whose tension 
is to be measured, communicates with the 
top of the cistern. If the mercury is at 
the same level in the cistern and tube, the 
pressure equals one atmosphere. As the 
tension in D increases, the column in A B 
rises, and compresses the air in the tube. 
The tension of the air in A B above the 

mercury, together with the weight of the mercury above the level 
in the cistern, is equal to the tension in D\ so that the number 
(2) will not be in the middle point between (1) 
and the top, but somewhat below. A scale of FIG. 187. 

atmospheres is calculated according to the pro 
portions of the instrument, and placed by the 
side of the tube. 

283. Apparatus for Preserving a Con 
stant Level. Let A B (Fig. 188) be a reser 
voir which supplies a liquid to the vessel C D ; 
and suppose it is desired to preserve the level at 
the point C in the vessel, while the liquid is dis 
charged from it irregularly or at intervals. This 
is accomplished by letting the discharge pipe E 
enter C D below the required level C, while the 
air is supplied to the reservoir only by a tube 
F B, which just reaches that level. So long as 



the liquid in D is below (7, it is 
at a greater depth from the sur 
face A than B is, and therefore the 
pressure is greater, and the liquid 
will run from E, and air enter at B. 
But when the vessel is filled to (7, 
the hydrostatic pressures at C and 
B are equal ; it is therefore impos 
sible that the water should over 
come the air at G and pass out, and 
that the air should at the same time 
overcome the water at B, and pass 
in. Hence, if G discharges more 
slowly than E, it is immaterial 
whether water is running from Gr 
or not; the vessel will remain al 
ways filled to the level G B. 

FIG. 188. 



284. Quantity of the Atmosphere. Since the air sus 
tains a column of mercury thirty inches high, the weight of the 
whole atmosphere is equal to that of a stratum of mercury thirty 
inches thick covering the globe. The thickness is relatively so 
small that the volume of the stratum may be reckoned as that of 
a parallelepiped, thirty inches in height, and having a base equal 
to the surface of the earth. 

Letting R = the radius of the earth, and h = the depth of 
mercury, the earth s surface 4 TT R*, and the volume of mercury 
= 4 77 R 9 Ji = 4 x 3.14159 x (3956 x 5280) 2 x 2.5 cubic feet. 

This multiplied by 62.5 x 13.6, the weight of a cubic foot of 
mercury, gives about 11,650,000,000,000,000,000 Ibs. This is, 
therefore, the weight of the earth s atmosphere. 

285. Virtual Height of the Atmosphere. When two 

fluid columns are in equilibrium with each other, their heights 
are inversely as their specific gravities (Art. 221). The specific 
gravity of mercury is 10464 times that of the air at the ocean 
level. Therefore, if the air had the same density in all parts, its 
height would be found by the proportion, 

1 : 10464 : : 2.5 : 26160 feet, 


which is almost five miles. Hence, the quantity of the entire at 
mosphere of the earth is pretty correctly conceived of when we 
imagine it having the density of that which surrounds us, and 
reaching to the height of five miles. 

286. Decrease of Density. But the atmosphere is very 
far from being throughout of uniform density. The great cause 
of inequality is the decreasing weight of superincumbent air at 
increasing altitudes. The law of diminution of density, arising 
from this cause, is the following! 

TJie densities of the air decrease in a geometrical as the altitudes 
increase in an arithmetical ratio. For, let us suppose the air to be 
divided into horizontal strata of equal thickness, and so thin that 
the density of each may be considered as uniform throughout. 
Let a be the weight of the whole column from the top to the 
earth, b the weight of the whole column above the lowest stratum, 
c that of the whole column above the second, &c. Then the 
weight of the lowest stratum is a b, and the weight of the 
second is b c, &c. Now the densities of these strata, and there 
fore their weights (since they are of equal thickness), are as the 
compressing forces ; or, 

a b:b c::b:cj 
.*. a c b c = & b c ; /. a c = V ; 


in the same way, b : c : c : d\ 

that is, the weights of the entire columns, from the successive 
strata to the top of the atmosphere, form a geometrical series; 
therefore, the densities of the successive strata, varying as the com 
pressing forces, also form a geometrical series. If, therefore, at a 
certain distance from the earth, the air is twice as rare as at the 
surface of the earth, at twice that distance it will be four times as 
rare, at three times that distance eight times as rare, &c. 

By barometric observations at different altitudes, it is found 
that at the height of three and a half miles above the earth the air 
is one-half as dense as it is at the surface. Hence, making an 
arithmetical series, with 3-J- for the common difference, to denote 
heights, and a geometrical series, with the ratio of J, to denote 
densities, we have the following: 

Heights, 3A, 7, 10, 14, m, 21, 24J, 28, 31$, 35. 
Densities, , \, , j g, 3 -^, g^, T ^ g , ^i g , g{ 7 , 7^4. 

According to this law, the air, at the height of 35 miles, is at 
least a thousand times less dense than at the surface of the earth. 
It has, therefore, a thousand times less weight resting upon it; in 
other words, only one-thousandth part of the air exists above that 


2B7. Actual Height of the Atmosphere. The foregoing 
law, founded on that of Mariotte, cannot, however, be applicable 
except to moderate distances. If it were strictly true, the atmo 
sphere would be unlimited. But that is impossible on a revolving 
body, since the centrifugal force must at some distance or other 
equal the force of gravity, and thus set a limit to the atmosphere ; 
and that limit in the case of the earth is more than 20,000 miles 
high. The actual height of the atmosphere is doubtless far below 
this ; for there can be none above the point where the repellency 
of the particles is less than their weight ; and the repellency di 
minishes just as fast as the density, while the weight diminishes 
very slowly. The highest portions concerned in reflecting the 
sunlight are about 45 miles above the earth. But there is reason 
to believe that the air extends much above that height, probably 
100 or 200 miles from the earth. 

288. The Motions of the Air. The air is never at rest. 
When in motion, it is called wind. The equilibrium of the atmo 
sphere is disturbed by the unequal heat on different parts of the 
earth. The air over the hotter portions becomes lighter, and is 
therefore pressed upward by the cooler and heavier air of the less 
heated regions. And the motions thus caused are modified as to 
direction and velocity by the rotation of the earth on its axis. 

289. The Trade Winds. The most extensive and regular 
system of winds on the earth is known by the name of the trade 
winds, so called on account of their great advantage to commerce. 
They are confined to a belt about equal in width to the torrid 
zone, but whose limits are four or five degrees further north than 
the tropics. 

In the northern half of this trade-wind zone the wind blows 
continually from the northeast, and in the southern half from the 
southeast. As these currents approach each other, they gradually 
become more nearly parallel to the equator, while between them 
there is a narrow belt of calms, irregular winds, and abundant 

The oblique directions of the trade winds are the combined 
effects of the heat of the torrid zone and the rotation of the earth. 
The cold air of the northern hemisphere tends to flow directly 
south, and crowd up the hot air over the equator. In like manner, 
the cold air of the southern hemisphere tends to flow directly 
northward. So that if the earth were at rest, there would be north 
winds on the north side of the equator, and south winds on the 
south side. But the earth revolves on its axis from west to east, 
and the air, as it moves from a higher latitude to a lower, has only 
so much eastward motion as the parallel from which it came. 


Therefore, since it really has a less motion from the west than 
those regions over which it arrives, it has relatively a motion from 
the east. This motion from the east, compounded with the motion 
from the north on the north side of the equator, and with that 
from the south on the south side, constitutes the northeast and 
southeast tradewinds. 

The limits of this system move a few degrees to the north dur 
ing the northern summer, and to the south during the northern 
winter, but very much less than might be expected from the 
changes in the sun s declination. 

In certain localities w r ithin the tropics the wind, owing to 
peculiar configurations of coast and elevations of the interior, 
changes its direction periodically, blowing six months from one 
point, and six months from a point nearly opposite. The monsoons 
of southern India are the most remarkable example. 

290. The Return Currents. The air which is pressed up 
ward over the torrid zone must necessarily flow away northward 
and southward towards the higher latitudes, to restore the equi 
librium. Hence, there are south winds in the upper air on the 
north side of the equator, and north winds on the south side. But 
these upper currents are also oblique to the meridians, because, 
having the easterly motion of the equator, they move faster than 
the parallels over which they successively arrive, so that a motion 
from the west is combined with the others, causing southwest 
winds in the northern hemisphere, and northwest in the southern. 
These motions of the upper air are discovered by observations 
made on high mountains, and in balloons, and by noticing the 
highest strata of clouds. It is to be borne in mind that although 
the atmosphere is more than 100 miles high, yet the lower half 
does not extend beyond three and a half miles above the earth 
(Art. 280). 

291. Circulation Beyond the Trade Winds. The upper 
part of the air which flows away from the equator cannot wholly 
retain its altitude, because of the diminishing space on the suc 
cessive parallels. About latitude 30, it is so much accumulated 
that it causes a sensible increase of pressure (Art. 267), and begins 
to descend to the earth. It is probable that some of the descend 
ing air still retains its oblique motion towards higher latitudes 
(for the prevailing winds of the northern temperate zone are from 
the southwest, and of the southern temperate zone from the north 
west), while a part joins with the lower air which is moving 
towards the equator. Only so much of the rising equatorial mass 
can flow back to the polar regions as is needed to supply the 
comparatively small area within them. On account of the sue- 


cessive descent of the air returning from the equator, there is 
much less distinctness and regularity in the general circulation 
outside of the torrid zone than within it. Besides this, various 
local causes, such as mountain ranges, sea-coasts, and ocean cur 
rents, clear and cloudy skies, &c., mingle their effects with the 
more general circulation, and modify it in every possible way. 

292. Land and Sea Breezes. These are limited circula 
tions over adjoining portions of land and water, the wind blowing 
from the water to the land in the day time, and in the contrary 
direction by night. When the sun begins to shine each day, it 
heats the land more rapidly than the water. Hence the air on 
the land becomes warmer and lighter than that on the water, and 
the surface current sets toward the land. By night the flow is re 
versed, because the land cools most rapidly, and the air above it 
becomes heavier than that over the water. These effects are more 
striking and more regular in tropical countries, but are common 
in nearly all latitudes. 

293. A Current Through a Medium. There are some 
phenomena relating to currents moving through a fluid, either of 
the same or a different kind, which belong alike to hydraulics and 
pneumatics ; a brief account of these is presented here. 

If a stream is driven through a medium, it carries along the 
adjoining particles by friction or adhesion. The experiment of 
Venturi illustrates this kind of action, as it takes place between 
the particles of water. A reservoir filled with water has in it an 
inclined plane of gentle ascent, whose summit just reaches the 
edge of the reservoir. A stream of water is driven up this plane 
with force sufficient to carry it over the top ; but in doing so, it 
takes out continually some part of the water of the reservoir, and 
will in time empty it to the level of the lowest part of the stream. 
A stream of air through air produces the same effect, as may be 
shown by the flame of a lamp near the stream always bending to 
ward it. In like manner, water through air carries air with it ; 
when a stream of water is poured into a vessel of water, air is car 
ried doAvn in bubbles ; and cataracts carry down much air, which 
as it rises forms a mass of foam on the surface. The strong wind 
from behind a high waterfall is owing to the condensation of air 
brought down by the back side of the sheet. 

294. Ventilators. If the stream passes across the end of an 
open tube, the air within the tube will be taken along with the 
stream, and thus a partial vacuum formed, and a current estab 
lished. It is thus that the wind across the top of a chimney in 
creases the draught within. To render this effect more uniformly 


successful, by preventing the wind from striking the interior edge 
of the flue, appendages, called ventilators, 
are attached to the chimney top. A sim- FIG. 189. 

pie one, which is generally effectual, con 
sists of a conical frustum surrounding the 
flue, as in Fig. 189, so that the wind, on 
striking the oblique surface, is thrown 
over the top in a curve, which is convex 
upward. The same mechanical contriv 
ance is much used for the ventilation of 
public halls and the holds of ships. A 
horizontal cover may be supported by rods, 
at the height of a few inches, to prevent 
the rain from entering. 

295. A Stream Meeting a Surface. Though the moving 
fluid may be elastic, yet, when it meets a surface, it tends to follow 
it, rather than to rebound from it. This effect is partly due to 
adhesion, and partly to the resistance of the medium in which the 
stream moves. It will not only follow a plane or concave surface, 
but even one which is convex, provided the velocity of the current 
is not too great, or the curvature too rapid. A stream of air, 
blown from a pipe upon a plane surface, will extinguish the flame 
of a lamp held in the direction of the surface beyond its edge, 
while, if the lamp be held elsewhere near the stream, the flame 
will point toward the stream, according to Art. 293. Hence, snow 
is blown away from the windward side of a tight fence, and from 
around trees. 

296 Diminution of Pressure on a Surface. When a 
stream is thus moving along a surface, the fluid pressure on that 
surface is slightly diminished. This is proved by many experi 
ments. If a curved vane be suspended on a pivot, and a stream 
of air be directed tangentially along the surface, it will move to 
ward the stream, and may be made to revolve rapidly by repeating 
the blast at each half revolution. What is frequently called the 
pneumatic paradox is a phenomenon of the same kind. A stream 
of air is blown through the centre of a disk, against another light 
disk, which, instead of being blown off, is forcibly held near to it 
by the means. The pressure is diminished by all the radial streams 
along the surface contiguous to the other disk, and the full press 
ure on the outside preponderates. Another form of the experi 
ment is to blow a stream of air through the bottom of a hemi 
spherical cup, in which a light sphere is lying loosely. The sphere 
cannot be blown out, but, on the contrary, is held in, as may be 
seen by inverting the cup, while the blast continues. It appears 


to be for a reason of the same sort that a ball or a ring is sustained 
by a jet of water. It lies not on the top, but on the side of the 
jet, which diminishes the pressure on that side of the ball, so that 
the air on the outside keeps it in contact. The tangential force 
of the jet causes the body to revolve with rapidity. A ball can be 
sustained a few inches high by a stream of air. 

297. Vortices where the Surface Ends. As a current 
reaches the termination of the surface along which it was flowing, 
a vortex or whirl is likely to occur in the surrounding medium 
behind the edge of the surface. Vortices are formed on water, 
whose flow is obstructed by rocks ; and often when the obstruct 
ing body is at a distance below the surface, the whirl which is es 
tablished there is communicated to the top, so that the vortex is 
seen, while its cause is out of sight. There is a depression at the 
centre, caused by the centrifugal force ; and if the rotation is 
rapid, a spiral tube is formed, in which the air descends to great 
depths. These are called whirlpools. In a similar manner whirls 
are produced in the air, when it pours off from a surface. The 
eddying leaves on the leeward side of a building in a windy day 
often indicate such a movement, though it may have no perma 
nency, the vortex being repeatedly broken up and reproduced. 

298. Vortices by Currents Meeting. But vortices are 
also formed by counteracting currents in an open medium. When 
an aperture is made in the middle of the bottom of a vessel, as the 
water runs toward it, the filaments encounter each other, and 
usually, though not invariably, they establish a rotary motion, 
and form a whirlpool. Vortices are a frequent phenomenon of 
the atmosphere, sometimes only a few feet in diameter, in other 
instances some rods or even miles in width. The smaller ones, 
occurring over land, are called whirlwinds ; over water, ivater- 
spouls. They probably originate in currents which do not exactly 
oppose each other, but act as a couple of forces, tending to produce 
rotation (Art. 54). 

The burning of a forest sometimes occasions whirlwinds, which 
are borne away by the wind, and maintain their rotation for miles. 
As the pressure in the centre is diminished by the centrifugal 
force, substances heavier than air, as leaves and spray, are likely to 
be driven up in the axis, and floating substances, as cloud, will for 
the same reason descend. The rising spray and the descending 
cloud frequently mark the progress of a vortex in the air, as it 
moves over a lake or the ocean. Such a phenomenon is called a 

For a notice of cyclones, see Part VIII, on Heat. 





299. Sound. Vibrations. The impression which the mind 
receives through the organ of hearing is called sound. But the 
same word is constantly used to signify that progressive vibratory 
movement in a medium by which the impression is produced, as 
when we speak of the velocity of sound. 

This is one of the several modes of motion mentioned in Art. 4. 
The vibrations constituting sound are comparatively slow, and are 
often perceived by sight and by feeling as well as by hearing. For 
these reasons, the true nature of sound is investigated with far 
greater ease than that of light, electricity, &c. It is not difficult 
to discover that vibrations in the medium about us are essential to 
hearing; and these vibrations are always traceable to the body in 
which the sound originates. A body becomes a source of sound 
by producing an impulse or a series of impulses on the surrounding 
medium, and thus throwing the medium itself into motion. A 
single sudden impulse causes & noise, with very little continuance; 
an irregular and rapid succession of impulses, a crash, or roar, or 
continued noise of some kind ; but if the impulses are rapid and 
perfectly equidistant, the effect is a musical sound. In most cases 
of the last kind the impulses are vibrations of the body itself; and 
whatever affects these vibrations is found to affect the sound em 
anating from it ; and if they are destroyed, the sound ceases. 

If we rub a moistened finger along the edge of a tumbler 
nearly full of water, or draw a bow across the strings of a viol, we 
can procure sounds which remain undiminished in intensity as 
long as the operation by which they are excited is continued. In 
both cases the vibrations are visible ; those of the tumbler are 
plainly seen as crispations on the water to which they are commu 
nicated ; the string appears as a broad shadowy surface. If a wire 
or light piece of metal rests against a bell or glass receiver, when 

190 SOUND. 

ringing, it will be made to rattle. If sand be strewed on a hori 
zontal plate while a bow is drawn across its edge, the sand will be 
agitated, and dance over the surface, till it finds certain places 
where vibrations do not exist. Near an organ-pipe the tremor of 
the air is perceptible, and pipes of the largest size jar the seats and 
walls of an edifice. Every species of sound may be traced to im 
pulses, or vibrations in the sounding body. 

300. Sonorous Bodies. Two qualities in a body are neces 
sary, in order that it may be sonorous. It must have a form 
favorable for vibratory movements, and sufficient strength of elas 

The favorable forms are in general rods and plates, rather than 
very compact masses, like spheres and cubes ; because the particles 
of the former are more free to receive lateral movements than 
those of the latter, which are constrained on every side. But even 
a thin lamina may have a form which allows too little freedom of 
motion, such as a spherical shell, in which the parts mutually sup 
port each other. If the shell be divided, the hemispheres are 
bell-shaped and very sonorous. 

The elasticity of some materials is too imperfect for continued 
vibration ; thus lead, in whatever form, has no sonorous quality. 
In other cases, where the elasticity is nearly perfect, yet it is a 
feeble force, and hence the vibrations are slow and inaudible. 
Thus india-rubber is quite elastic, but its force is feeble, and occa 
sions but little sound. 

301. Air as a Medium of Sound. There must not only 
be a vibrating body, as a source of sound, but a medium for its 
communication to the organ of hearing. The ordinary medium is 
air. Let a bell mounted with a hammer and mainspring, so as to 
continue ringing for several minutes, be placed on a thick cushion 
under the receiver of an air-pump. The cushion, made of several 
thicknesses of woolen cloth, is necessary to prevent communica 
tion through the metallic parts of the instrument. As the pro 
cess of exhaustion goes on, the sound of the bell grows fainter, 
and at length ceases entirely. From this experiment we learn that 
sound cannot be propagated through a vacant space, even though 
it be only an inch or two in extent ; and also that air conveys 

sound more feebly as it is more rare. The latter is proved by the 
faintness of sounds on the tops of high mountains. Travelers 
among the Alps often observe that at great elevations a gun can 
be heard only a small distance. The fact that meteoric bodies are 
sometimes heard when passing over at the height of 40 or 50 miles 
does not conflict with the above statements ; for the velocity of 
meteors is vastly greater than any other velocities which occur 


within the earth s atmosphere. On the other hand, when air has 
more than the natural density, it conveys sound with more inten 
sity, and therefore to a greater distance. In a diving-bell sunk to 
a considerahle depth a whisper is painfully loud. 

302. Velocity of Sound in Air. Sound occupies an ap 
preciable time in passing through air. This is a fact of common 
observation. The flash of a distant gun is seen before the report 
is heard. Thunder usually follows lightning after an interval of 
many seconds; but if the electric discharge is quite near, the 
lightning and thunder are almost simultaneous. If a person is 
hammering at a distance, the perceptions of the blows received by 
the eye and the ear do not generally agree with each other : or if 
in any case they do agree, it will be observed that the first stroke 
seen is inaudible, and the last one heard is invisible ; for it re 
quires just the time between two strokes for the sound of each to 
reach us. Many careful experiments were made in the 18th cen 
tury to determine the velocity of sound ; but as the temperature 
was not recorded, they have but little value. During the present 
century, the velocity has been determined by several series of ob 
servations in different countries, and all reduced for temperature 
to the freezing-point. The agreement between them is very close, 
and the mean of all is 1090 feet per second at 32 F. 

303. Velocity as Affected by the Condition of the Air 
and the Quality of the Sound. 

Temperature affects the velocity of sound; the latter is in 
creased about one foot (0.96 ft.) for each degree of rise in the tem 
perature. Therefore, in most Xew England climates, the velocity 
of sound varies more than 100 feet during the year on this ac 
count. Probably the celebrated experiments of Derham, in Lon 
don, 1708, who made the velocity 1142 feet, were performed in the 
heat of summer. 

Wind of course affects the velocity of sound by the addition or 
subtraction of its own velocity, estimated in the same direction, 
because it transfers the medium itself in which the sound is con 
veyed. This modification, however, is only slight, for sound 
moves ten times faster than wind in the most violent hurricane. 

But other changes in the condition of the air produce little or 
no effect. Neither pressure, nor moisture, nor any change of 
weather, alters the velocity of sound, though they may affect its 
intensity, and therefore the distance at which it can be heard. 
Falling snow and rain obstruct sound, but do not retard it. 

All kinds of sound the firing of a gun the blow of a ham 
mer the notes of a musical instrument, or of the voice, however 
high or low, loud or soft, are conveyed at the same rate. That 

192 SOUND. 

sounds of different pitch are conveyed with the same velocity 
was conclusively proved by Biot, in Paris, who caused several airs 
to be played on a flute at one end of a pipe more than 3000 feet 
long, and heard the same at the other end distinctly, and without 
the slightest displacement in the order of notes, or intervals of 
silence between them. 

304. The Calculated Velocity. For several years there 
was a large unexplained difference between the calculated velocity 
of sound and the actual velocity as determined by experiment. 
While the latter is, as already stated, 1090 feet per second at the 
freezing-point, calculation gave 916 feet. The difference was at 
length explained by La Place, who ascertained that it arises from 
the heat developed in the air by the compression which it under 
goes. The calculations previously made regarded the elasticity as 
varying with the density alone, according to Mariotte s law, as 
suming that the temperature remained unchanged. But it is a 
wejl-known fact that when air is compressed, a part of its latent 
heat becomes sensible, and raises its temperature. If the conden 
sation is gradual, the heat is radiated or conducted off, especially 
if in contact with other bodies ; but the heat developed in the 
propagation of sound has little opportunity to escape, and, though 
without continuance, it augments the elasticity of the air, so as to 
add 174 feet to the velocity of sound in it. 

305. Diffusion of Sound. Sound produced in the open air 
tends to spread equally in all directions, and will do so whenever 
the original impulses are alike on every side. But this is rarely 
the case. In firing a gun, the first impulse is given in one direc 
tion, and the sound will have more intensity, and be heard further 
in that direction than in others. It is ascertained by experiment, 
that a person speaking in the open air can be equally well heard 
at the distance of 100 feet directly before him, 75 feet on the right 
and left, and 30 feet behind him; and therefore an audience, in 
order to hear to the best advantage, should be arranged within 
limits having these proportions. But, as will be seen hereafter, 
this rule is not applicable to the interior of a building. 

Sound is also heard in certain directions with more intensity, 
and therefore to a greater distance, if an obstacle prevents its dif 
fusion in other directions. On one side of an extended wall sound 
is heard further than if it spread on both sides ; still further, in an 
angle between two walls ; and to the greatest distance of all, when 
confined on four sides, and limited to one direction, as in a long 
tube. The reason in these several cases is obvious ; for a given 
force can produce a given amount of motion ; and if the motion is 
prevented from spreading to particles in some directions, it will 


ivach more ^distant ones in those directions in which it does spread. 
Speaking-tubes confine the movement to a slender column of air, 
and therefore convey sound to great distances, and are on this ac 
count very useful in transmitting messages and orders between re 
mote parts of manufacturing edifices and public houses. 

306. Nature of Acoustic Waves. The vibrations of a 
medium in the transmission of sound are of the kind called longi 
tudinal ; that is, the particles vibrate longitudinally with regard 
to the movement of the sound ; whereas, in water-waves, the par 
ticle-motion is partly transverse to the wave-motion (Art. 247). 
If, for example, sound is passing from A to B (Fig. 190), the par- 

FIG. 190. 
r c r c r c 

tides just about A are (at tlu moment represented) in a state of 
condensation ; around this condensed centre is a rarefied portion, 
then a condensed portion, &c., as marked by the letters r, c, r , c\ 
r", &c. From r to c the particles are advancing; so likewise from 
r to c , and from r" to c". But from c to r , from c to r", &c., 
they are rebounding. The condensed wave near B has advanced 
from A, and others have followed it at equal intervals; and be 
tween these waves of condensation are waves of rarefaction, which 
in like manner spread outward from the centre A. And yet no 
one particle has any other motion than a small vibration back and 
forth in the line, near its original place of rest. The amplitude is 
the distance through which a particle vibrates. The intensity or 
loudness of sound depends on the amplitude. 

In water-waves we distinguish carefully between the motion 
of the wave and the motion of the water which forms the wave ; 
so here, the wave-motion is totally different from the motion of 
the air itself. The wave, i. e. the state of condensation and subse 
quent rarefaction, travels swiftly forward ; but the masses of air, 
which suffer these condensations and rarefactions, simply tremble 
in the line of that motion. 

Since the motion is propagated in all directions alike, the en 
tire system of waves around the point where sound originates con 
sists of spherical strata of air alternately condensed and rarefied. 
As the quantity set in motion in these successive layers increases 

194 SOUND. 

with the square of the distance, the amount of motion communi 
cated to each particle must diminish in the same ratio. Hence, 
the intensity of sound varies inversely as the square of the dis 

Fig. 190 is a section of a system of spherical waves around the 
source A. 

A ray of sound is any one of the radii of the sphere whose 
centre is the source of sound. The vibratory motion is propagated 
along each of the rays. 

307. Other Gaseous Bodies, as Media of Sound. Let 

a spherical receiver, having a bell suspended in it, be exhausted of 
air, till the bell ceases to be heard ; then fill it with any gas or va 
por instead of air, and the bell will be heard again. By means of 
an organ-pipe blown by different gases, it can be learned with 
what velocity sound would move in each kind of gas experimented 
upon, because the pitch of a given pipe depends upon the velocity 
of the waves, as will be seen hereafter. In hydrogen sound is ex 
ceedingly feeble, but moves nearly three times as fast as in air. 
Momentary development of heat by compression produces, in all 
gaseous bodies, the effect of increasing the velocity of sound. 

308. Liquids as Media. Many experimenters have deter 
mined the circumstances of the propagation of sound in water. 
Franklin found that a person with his head under water could 
hear the sound of two stones struck together at a distance of more 
than half a mile. In 1826, Colladon made many careful experi 
ments in the water of Lake Geneva. The results of these and 
other trials are principally the following : 

1. Sounds produced in the air are very faintly heard by a per 
son in water, though quite near ; and sounds originating under 
water are feebly communicated to the air above, and in positions 
somewhat oblique are not heard at all. 

2. Sounds are conveyed by water with a velocity of 4700 feet 
per second, at the temperature of 47 F., which is more than four 
times as great as in air. The calculated and the observed velocity 
of sound in water agree so nearly with each other, that there ap 
pears to be no appreciable effect arising from heat developed by 

3. Sounds conveyed in water to a distance, lose their sonorous 
quality. For example, the ringing of a bell gives a succession of 
short sharp strokes, like the striking together of two knife-blajdes. 
The musical quality of the sound is noticeable only within 600 or 
700 feet. In air, it is well known that the contrary takes place ; 
the blow of the bell-tongue is heard near by, but the continued 
musical note is all that affects the ear at a distance. 


4. Acoustic shadows are formed; that is, sound passes the 
edges of solid bodies nearly in straight lines, and does not turn 
around them except in a very slight degree. In this respect, 
sound in water resembles light much more than it does sound in 

To enable the experimenter to hear distant sounds without 
placing himself under water, Colladon pressed down a cylindrical 
tin tube, closed at the bottom, thus allowing the acoustic pulses in 
the water to strike perpendicularly on the sides of the tube. In 
this way, the faintest sounds were brought out into the air. It 
appears to be true of sound as of light, that it cannot pass from a 
denser to a rarer medium at large angles of incidence, but suffers 
nearly a total reflection. 

309. Solids as Media. Solid bodies of high elastic energy 
are the most perfect media of sound which are known. An iron 
rod as, for instance, a lightning-rod will convey a feeble sound 
from one extremity to the other, with much more distinctness 
than the air. If the ears are stopped, and one end of a long wire 
is held between the teeth, a slight scratch or blow on the remote 
end will sound very loud. The sound in this case travels through 
the wire and the bones of the head to the organ of hearing. The 
stethoscope, an instrument used by physicians for determining 
whether the lungs or heart have a diseased or healthy action, illus 
trates the conduction of sound by solids. The instrument is a 
tubular rod of wood, one end of which is pressed upon the chest 
of the patient, while the ear is applied to the other. The move 
ments of the vital organs are thus distinctly heard, and the char 
acter of those movements readily distinguished. The sound of 
earthquakes and volcanic eruptions is transmitted to great dis 
tances through the solid earth. By laying the ear to the ground, 
the tramp of cavalry may be heard at a much greater distance 
than through the air. 

310. Velocity in Solids. Structure. The velocity of 
sound in cast iron is about 11000 feet per second ten times 
greater than in air. This was determined by Biot, in his experi 
ments on some aqueduct pipes in Paris, already alluded to. A 
blow upon one end was brought to an observer at the other 
through two channels, and seemed to be two blows. One sound 
traveled in the air within the tube, the other in the iron itself of 
which the pipe was made. From the observed interval of time 
between the two sounds, and the known velocity of sound in air, 
the velocity in iron is readily calculated. The pitch of sound pro 
duced by rods and tubes of different materials, when vibrating 

196 SOUND. 

longitudinally, enables us to determine with tolerable accuracy the 
velocity of propagation in those substances respectively. 

In one important particular solids differ from fluids, namely, 
in the fixed relations of the particles among themselves. These 
relations are usually different in different directions ; hence, sound 
is likely to be transmitted more perfectly in some directions 
through a given solid than in others. The scratch of a pin at 
one end of a stick of timber seems loud to a person whose ear is at 
the other end. The sound is heard more perfectly in the direction 
of the grain than across it. In crystallized substances it is unques 
tionably true that the vibrations of sound move with different 
speed and with different intensity in the line of the axis, and in a 
line perpendicular to it. 

311. Mixed Media. In all the foregoing statements it has 
been supposed that the medium was homogeneous; in other 
words, that the material, its density, and its structure, continue 
the same, or nearly the same, the whole distance from the source 
of sound to the ear. If abrupt changes occur, even a few times, 
the sound is exceedingly obstructed in its progress. When the re 
ceiver is set over the bell on the pump plate, the sound in the 
room is very much weakened, though the glass may not be one- 
eighth of an inch in thickness, and is an excellent conductor of 
sound. The vibrations of the internal air are very imperfectly 
communicated to the glass, and those received by the, glass pass 
into the air again with a diminished intensity. If a glass rod ex 
tended the whole distance from the bell to the ear, the sound 
would arrive in less time, and with more loudness, than if air 
occupied the whole extent. For a like reason, walls, buildings, or 
other intervening bodies, though good conductors of sound them 
selves, obstruct the progress of sound in the air. This explains 
the fact mentioned in Art. 308, that sound in air is heard faintly 
in water, and vice versa. When the texture of a substance is 
loose, having many alternations of material, it thereby becomes 
unfit for transmitting sound. It is for this reason that the bell- 
stand, in the experiment just referred to, is set on a cushion made 
of several thicknesses of loose flannel, that it may prevent the vi 
brations from reaching the metallic parts of the pump. The waves 
of sound, in attempting to make their way through such a sub 
stance, continually meet with new surfaces, and are reflected in all 
possible directions, by which means they are broken up into a 
multitude of crossing and interfering waves, and are mutually de 
stroyed. A tumbler, nearly filled with water, will ring clearly ; 
but if filled with an effervescing liquid, it will lose all its sonorous 
quality, for the same cause as before. The alternate surfaces of the 
liquid and gas, in the foam, confuse the waves, and deaden the sound. 



FIG. 191. 



312. Reflection of Sound. Sound is reflected from surfaces 
in accordance with the common law of reflection in the case of 
elastic bodies; that is, 

The angle of incidence equals the angle of reflection, and the two 
angles are on opposite sides of the perpendicular to the reflecting 

Suppose sound to emanate from A (Fig. 191), and meet the 
plane surface B D. The particles of air in the ray A E vibrate 
back and forth in that line, 
and those contiguous to B 
will, after striking the sur 
face, rebound on the line 
B G, as an elastic ball would 
do (Art. 103), and propa 
gate their motion along 
that line. The angle of 
incidence A B ^equals the 
angle of reflection F B G, 
and the two angles are on 
opposite sides ofFB, which 
is perpendicular to the re 
flecting surface B D. If 
G B be produced back 
ward, it will meet the per 
pendicular A E at C t as 

far behind B D as A is before it. In like manner, every ray of 
sound after reflection proceeds as if from (7, and the successive 
waves are situated as represented by the dotted lines in the figure. 
From the point E the reflection is directly back in the line E A. 

313. Echoes. When sound is so distinctly reflected from a 
surface that it seems to come from another source, it is called an 
echo. Broad and even surfaces, such as the walls of buildings and 
ledges of rock, often produce this effect. According to the law 
(Art. 312), a person can hear the echo of his own voice only by 
standing in a line which is perpendicular to the echoing surface. 
In order that one person may hear the echo of another s voice, 
they must place themselves in lines making equal angles with the 

198 SOUND. 

The interval of time between a sound and its echo enables one 
to judge of the distance of the surface, since the sound must -pass 
over it twice. Thus, if at the temperature of 74 the echo of the 
speaker s voice reaches him in two seconds after its utterance, the 
distance of the reflecting body is about 1130 feet, and in that pro 
portion for other intervals. And he can hear a distinct echo of 
as many syllables as he can pronounce while sound travels twice 
the distance between himself and the echoing surface. 

314. Simple and Complex Echoes. When a sound is 
returned by one surface, the echo is called simple ; it is called 
complex when the reflection is from two or more surfaces at differ 
ent distances, each surface giving one echo. Thus, a cannon fired 
in a mountainous region is heard for a long time echoed on all 
sides, and from various distances. 

A complex echo may also be produced by two parallel walls, if 
the hearer and the source of sound are both situated between 
them. The firing of a pistol between parallel walls a few hundred 
feet apart has been known to return from 30 to 40 echoes before 
they became too faint to be heard. The rolling of thunder is in 
part the effect of reverberation between the earth and the clouds. 
This is made certain by the observed fact that the report of a can 
non, which in a level country and under a clear sky is sharp and 
single, becomes in a cloudy day a prolonged roar, mingled with 
distant and repeated echoes. But the peculiar inequalities in the 
reverberations of thunder are doubtless due in part to the irregu 
larly crinkled path of the electric spark. A discharge of lightning 
occupies so short a time, that the sound may be considered as 
starting from all points of its track at once. But that track is 
full of large and small curves, some convex and some concave to 
the ear, and at a great variety of distance ; and all points which 
are at equal distances would be heard at once. Hence, the origi 
nal sound comes to the hearer with great irregularity, loud at one 
instant and faint at another. These inequalities are prolonged 
and intensified by the echoes which take place between the clouds 
and the earth. , 

315. Concentrated Echoes. The divergence of sound from 
a plane surface continues the same as before, that is, in spherical 
waves, whose centre is at the same distance behind the plane as 
the real source is in front. But concave surfaces in general pro 
duce a concentrating effect. A sound originating in the centre 
of a hollow sphere will be reflected back to the centre from every 
point of the surface. If it emanates from one focus of an ellipsoid, 
it will, after reflection, all be collected at the other focus. So, if 
two concave paraboloids stand facing each other, with their axes 


coincident, and a whisper is made at the focus of one, it will be 
plainly heard at the focus of the other, though inaudible at all 
points between. In the last case the sound is twice reflected, and 
passes from one reflector to the other in parallel lines. All these 
effects are readily proved from the principle that the angles of in 
cidence and reflection are equal. 

The speaking-trumpet and the ear- trumpet have been supposed 
by many writers to owe their concentrating power to multiplied 
reflections from the inner surface. But a part of the effect, and 
sometimes the whole, is doubtless due to the accumulation of force 
in one direction, by preventing lateral diffusion, till the intensity 
is greatly increased. 

Concave surfaces cause all the curious effects of what are called 
whispering galleries, such as the dome of St. Paul s, in London. 
In many of these instances, however, there seems to be a contin 
ued series of reflections from point to point along the smooth con 
cave wall, which all meet simultaneously (if the curves are of equal 
length) at the opposite point of the dome ; for the whisperer 
places his mouth, and the hearer his ear, close to the wall, and not 
in a focus of the curve. The Ear of Dionysius was probably a 
curved wall of this kind in the dungeons of Syracuse. It is said 
that the words, and even the whispers, of the prisoners were gath 
ered and conveyed along a hidden tube to the apartment of the 
tyrant. The sail of a ship when spread, and made concave by the 
breeze, has been known to concentrate and render audible to the 
sailors the sound of a bell 100 miles distant. A concave shell held 
to the ear concentrates such sounds as may be floating in the air, 
and is suggestive of the murmur of the ocean. 

316. Resonance of Rooms. If a rectangular room has 
smooth, hard walls, and is unfurnished, its reverberations will be 
loud and long-continued. Stamp on the floor, or make any other 
sudden noise, and its echoes passing back and forth will form a 
prolonged musical note, whose pitch will be lower as the apart 
ment is larger. This is called the resonance of the room. Now, 
let furniture be placed around the walls, and the reverberations 
will be weakened and less prolonged. Especially will this be the 
case if the articles be of the softer kinds, and have irregular sur 
faces. Carpets, curtains, stuffed seats, tapestry, and articles of 
dress have great influence in destroying the resonance of a room. 
The appearance of an apartment is not more changed than is its 
resonance by furnishing it with carpet and curtains. The blind, 
on entering a strange room, can, by the sound of the first step, 
judge with tolerable accuracy of its size and the general character 
of its furniture. 

200 SOUND. 

The reason why substances of loose textnre do not reflect sound 
well, is essentially the same as what has been stated (Art 311) for 
their not transmitting well ; they are not homogeneous the waves 
are reflected in all directions by successive surfaces, interfere with 
each other, and are destroyed. 

317. Halls for Public Speaking. In large rooms, such as 
churches and lecturing halls, all echoes which can accompany the 
voice of the speaker syllable by syllable, are useful for increasing 
the volume of sound ; but all which reach the hearers sensibly 
later, only produce confusion. It is found by experiment that if 
a sound and its echo reach the ear within one-sixteenth of a second 
of each other, they seem to be one. Hence, this fraction of time 
is called the limit of perceptibility. Within that time an echo can 
travel about 70 feet more than the original sound, and yet appear 
to coincide with it. If an echoing wall, therefore, is within 35 
feet of the speaker, each syllable and its echo will reach eveiy 
hearer within the limit of perceptibility. The distance may, how 
ever, be increased to 40 or even 50 feet without injury, especially 
if the utterance is not rapid. Walls intended to aid by their 
echoes should be smooth, but not too solid; plaster on lath is 
better than plaster on brick or stone ; the first echo is louder, and 
the reverberations less. Drapery behind the speaker deprives him 
of the aid of just so much echoing surface. A lecturing hall is 
improved by causing the wall behind the speaker to change its di 
rection, on the right and left of the platform, at a very obtuse 
angle, so as to exclude the rectangular corners from the room. 
The voice is in this way more reinforced by reflection, and there 
is less resonance arising from the parallelism of opposite walls. 
Paneling, and any other recesses for ornamental purposes, may 
exist in the reflecting walls without injury, provided they are not 
curved. The ceiling should not be so high that the reflection 
from it would be delayed beyond the limit of perceptibility. Con 
cave surfaces, such as domes, vaults, and broad niches, should be 
carefully avoided, as their effect generally is to concentrate all the 
sounds they reflect. An equal diffusion of sound throughout the 
apartment, not concentration of it to particular points, is the ob 
ject to be sought in the arrangement of its parts. 

As to distant parts of a hall for public speaking, the more com 
pletely all echoes from them can be destroyed, the more favorable 
is it for distinct hearing. It is indeed true that if a hearer is with 
in 35 feet of a wall, however remote from the speaker, he will hear 
a syllable, and its echo from that wall, as one sound ; but to all 
the audience at greater distances from the same wall, the echoes 
will be perceptibly retarded, and fall upon subsequent syllables, 


thus destroying distinctness. The distant walls should, by some 
means, be broken* up into small portions, presenting surfaces in 
different directions. A gallery may aid in effecting this ; and the 
seats of the gallery and of the lower iloor may rise rapidly one be 
hind another, so that the audience will receive directly much of the 
sound which would otherwise go to the remote wall, and be re 
flected. Especially should no large and distant surfaces be paral 
lel to nearer ones, since it is between parallel walls that prolonged 
reverberation occurs. 

318. Refraction of Sound. It has been ascertained by ex 
periment that sound, like light, may be refracted, or bent out of 
its rectilinear course by entering a substance of different density. 
If a large convex lens be formed of carbonic acid gas, by inclosing 
it in a sphere of thin india-rubber, a feeble sound, like the ticking 
of a watch, produced on one side, will be concentrated to a focal 
point on the other. In this case, the several diverging rays of 
sound are refracted toward eacli other on entering the sphere, and 
still more on leaving it, so that they are converged to a focus. 

319. Inflection of Sound. If air- waves are allowed to pass 
through an opening in an obstructing wall, they are not entirely 
confined within the radii of the wave-system produced through 
the opening, but spread with diminished intensity in lateral direc 
tions. The particles near 

the edges of the opening, 
as B and C (Fig. 192) may 
be considered as sources 
of sound ; and if they be 
made centres of concentric 
spheres, whose radii are 
equal to the length of the 
wave, B b, or C c, and its 
multiples, then these spher 
ical surfaces will represent 
the lateral systems of 

waves which are diffused on every side of the direct beam, B D, 
C E. But the sound is in general more feeble as the distance 
from B D, or C E, is greater, and in certain points is destroyed by 
interference. This spreading of sound in lateral directions is 
called the inflection of sound. 

What is true of all sides of an opening is of course true when 
ever sound passes by the side of an obstacle. Instead of being 
limited by lines almost straight drawn from the source, as light is 
in the formation of a shadow, it bends round the edge, and is 

FIG. 192. 

202 SOUND. 

heard, though more feebly, behind the intervening body. It has 
been already noticed (Art. 308) that in water there is little or no 
inflection of sound. 



320. The Vibrations in Musical Sounds. When the im 
pulses of a sounding body upon the air are equidistant, and of suf 
ficient frequency, they produce what is termed a musical sound. 
In most cases these impulses are the isochronous vibrations of the 
body itself, but not necessarily so ; it is found by experiment that 
blows or pulses, of any species whatever, if they are more than 
about 15 or 20 per second, and possess the property of isoclironism, 
cause a musical tone. For example, the snapping of a stick on 
the teeth of a metallic wheel would seem as unlikely as anything 
to produce a musical sound; but when the wheel is in rapid 
motion, the succession causes a pure musical note. Equidistant 
echoes often produce a musical sound, as when a person stamps on 
the floor of a rectangular room, finished, but unfurnished (Art. 
316). So, on a walk by the side of a long baluster fence, a sudden 
sharp sound, like the blow of a hammer on a stone, brings back a 
tone more or less prolonged, resembling the chirp of a bird. It is 
occasioned by successive equidistant echoes from the balusters of 
the fence. A flight of steps will sometimes produce the same 
effect, the tone being on a lower key than that from the fence, as 
it should be. 

321. The Pitch of Musical Sounds. What is called the 
pitch of a musical sound, or its degree of acuteness, is owing en 
tirely to its rate of vibration. Other qualities of sounds are due 
to other and often unknown circumstances ; but rapidity of vibra 
tion is the only condition on which the pitch depends. In compar 
ing one musical sound with another, if the number of vibrations 
per second is greater, the sound is more acute, and is said to be of 
a higher pitch ; if the vibrations are fewer per second, the sound 
is graver, or of a lower pitch. 

322. The Monochord. If a string of uniform size and text 
ure is stretched on a box of thin wood, by means of a pulley and 
weight, the instrument is called a monochord, and is useful for 
studying the laws of vibrations in musical sounds. The sound 


emitted by the vibrations of the whole length of the string is 
called its, fundamental sound. 

If the string be drawn aside from its straight position, and 
then released, one component of the force of tension urges every 
particle back towards its place of rest ; but the string passes be 
yond that place, on account of the momentum acquired, sind de 
viates as far on the other side; from which position it returns, for 
the same reason as before, and continues thus to vibrate lill ob 
structions destroy its motion. By the use of a bow, the vibrations 
may be continued as long as the experimenter chooses. 

The pitch of the fundamental sound of musical strings is found 
by experience to depend on three circumstances; the length of the 
string its weight or quantity of matter and its tension. The 
tone becomes more acute as we increase the tension, or diminish 
either the length or the weight. The operation of these several 
circumstances may be seen in a common violin. The pitch of any 
one of the strings is raised or lowered by turning the screw so as 
to increase or lessen its tension ; or, the tension remaining the 
same, higher or lower notes are produced by the same string, by 
applying the fingers in such a manner as to shorten or lengthen 
the string which is vibrating; or, both the tension and the length 
of the string remaining the same, the pitch is altered by making 
the string larger or smaller, and thus increasing or diminishing its 

A string is said to make a single vibration in passing from the 
extreme limit on one side to the extreme limit on the other ; a 
double vibration is the motion across and back again to the origi 
nal position. Independently of calculation, it is easy to see that, 
with a given weight per inch, and a given tension, the string will 
vibrate slower, if longer, since there is more matter to be moved, 
and only the same force to move it ; and for a similar reason, the 
length and tension being given, it will also vibrate slower, if 
heavier. On the other hand, if length and weight are given, it 
will vibrate faster, if the tension is greater ; because a greater force 
will move a given quantity at a swifter rate. 

323. Time of a Single Vibration. The mathematical 
formula for the time of a vibration is the following, in which T 
the time of a single vibration ; I = the length of the string in 
inches ; w = the weight of one inch of the string ; / = the ten 
sion in Ibs. ; and g = the force of gravity = 386 inches = 32 
feet (Art. 28) ; 

204 SOUND. 

The constant factor, g, being omitted, the variation may be ex 
pressed thus : 

T oc =- ; that is, 

The time of a vibration varies as the length of the string multi 
plied by the square root of its weight per inch, and divided by the 
square root of its tension. 

As the distance of the string from its quiescent position does 
not form an element of the algebraic expression for the time of a 
vibration, it follows that the time is independent of the amplitude. 
Hence, as in the pendulum, the vibrations of a string, fixed at 
both ends, are performed in equal times, whether the amplitude 
of the vibrations be greater or smaller. It is on this account that 
the pitch of a string does not alter, when left to vibrate till it 
stops. The excursions from side to side grow less, and therefore 
the sound more feeble, till it ceases ; but the rate of vibration, and 
therefore the pitch, remains the same to the last. This property 
of isochronism, independent of extent of excursion, is common to 
sounding bodies generally, and is owing to what may be called the 
law of elasticity, that the restoring force, acting on any particle, 
varies directly as its distance from the place of rest. For example, 
each particle of the string, if removed twice as far from its place 
of rest, is urged back by a force twice as great, and therefore re 
turns in the same time. 

324. The Number of Vibrations in a Given Time. The 

greater is the length of one vibration, the less will be the number 
of vibrations in a given time ; that is, if N represents the number, 

N cc -=- : but as T <x , .. N oc . If t and w are con- 
T Vt iVw 

stant, N x -; if I and t are constant, N oc -7; and if I and w 
l V w 

are constant, N oc Vi; that is, 

1. The number of vibrations varies inversely as the length. 

2. The number of vibrations varies inversely as the square root 
of the weight of the string. 

3. The number of vibrations varies as the square root of the 

Thus, the number of vibrations in a second may be doubled, 
either by halving the length of the string or by making its weight 
one-fourth as great, or, finally, by making its tension four times as 

325. Vibrations of a String in Parts. The monochord 
may be made to vibrate in parts, the points of division remaining 


at rest ; and this mode of vibration may even coexist with the one 
already described. Of course the sound produced by the parts 
will be on a higher pitch, since they are shorter, while the tension 
and the weight per inch remain unaltered. It is a noticeable fact 
that the parts are always such as will exactly measure the whole 
without a remainder. Hence the vibrating parts are either halves, 
thirds, fourths, or other aliquot portions. The sounds produced 
by any of these modes of vibration are called harmonics, for a rea 
son which will appear hereafter. Suppose a string (Fig. 193) to 

FIG. 193. 

be stretched between A and B, and that it is thrown into vibra 
tion in three parts. Then while A D makes its excursion on ono 
side, D C will move in the opposite direction, and C B the same 
as A D ; and when one is reversed, the others are also, as shown 
by the dotted line. In this way D and G are kept at rest, being 
urged toward one side by one portion of string, and toward the 
opposite by the next portion. But the string may at the same 
time vibrate as a whole ; in which case D and C will have motion 
to each side of their former places of rest, while relatively to them 
the three portions will continue their movements as before. The 
points C and D are called nodes ; the parts A D, D C, and C B, 
are called ventral segments. By a little change in the quickness 
of the stroke, the bow may be made to bring from the monochord 
a great number of harmonic notes, each being due to the vibra 
tions of certain aliquot parts of the string. By confining a partic 
ular point, however, at the distance of 4, i, or other simple fraction 
of the whole from the end, the particular harmonic belonging to 
that mode of division may be sounded clear, and unmingled with 
the others. 

326. Vibrations cf a Column of Air. When a musical 
sound is produced by a pipe of any kind, it is the column of in 
closed air which must be regarded as the sounding body. A con 
densed wave is caused, by some mode of excitation, to travel back 
and forth in the pipe, followed by a rarefied portion ; and these 
waves affect the surrounding air much in the same way as do the 
alternate excursions of a string. That it is the air, and not the 
pipe itself, which is the source of sound, is proved by using pipes 
of various materials the most elastic and the most inelastic as 
prlass, wood, paper, and lead ; if they are of the same form and size, 
the tone in each case has the same pitch. 

In order to examine the manner in which the air-columns in 



FIG. 194. 

pipes perform their v brations, it is convenient to consider them 
in three classes : 

1st. Pipes which are closed at both ends. 

2d. Those which are closed at one end and open at the other. 

3d. Those which are open at both ends. 

327. Both Ends of the Pipe Closed. Suppose the ends 
of the pipe, AC B (Fig. 194), to be closed, and an impulse in some 
way to be communicated 

at the centre, (7; then the ====- 

motion of the column will [T 5 " 

consist of a constant and I 

regular fluctuation of the J_ c JS 

whole mass to and fro with 
in the pipe, the air being always condensed in one half, while it is 
rarefied in the other. While the condensed pulse moves from B 
to C t the point of rarefaction runs from A to (7, where they pass 
each other; hence, at the middle of the pipe there is no change of 
density, since every degree of condensation is at that point met by 
an equal degree of rarefaction of the other half of the general 
wave. At the extremities, A and B, there is alternately a maxi 
mum of condensation and of rarefaction, each being reflected and 
returning, to meet again at C. Fig. 195 shows the air in a state 
of condensation at A, and 

of rarefaction at B. At _ _Fre.l95. 

all points between the cen 
tre and the ends there is 
alternate condensation and 
rarefaction, but in a less 
degree according to the distance from the ends. 

On the other hand, the excursions of the particles are greatest 
at (7, and nothing at A and B, where all motion is prevented by 
the fixed stoppers by which the pipe is closed. Between the ends 
and the centre, the amplitude of vibration is greater, as the dis 
tance from the centre is less. 

The pitch of such a pipe will be lower, as the pipe is longer, 
because the waves have a greater distance to travel between the 
successive reflections, and hence there will be a smaller number 
per second. So also, lowering the temperature lowers the pitch, 
since the wave then travels more slowly, and suffers fewer reflec 
tions in a second. 

328. One End of the Pipe Closed, the other Open If, 

while the column A B is vibrating as a whole, an aperture is made 
at the centre, or even if the pipe is divided there, so that the aper 
ture extends entirely round it, this will not interrupt the oscilla- 


tion already described, because there is neither rarefaction nor 
condensation at the point G y and hence no tendency there to lat 
eral motion. The means employed for ex 
citing vibrations may therefore be applied FIG. 190. ^ 
at the open section. Let the pipe A B (Fig. 
196), remaining stopped at A and B, be di 
vided at the middle, C\ and let the half 
pipe B C be removed, while the exciting 

cause remains at C (Fig. 19-i), then the vibrations in A C will still 
continue, and the pitch be unaltered. For now the condensed 
pulse, on reaching C, will be returned to A by the vibrating disk 
or spring which excites it, and will make a second reflection at A 
at the same instant as it would have done at B in the whole pipe 
A B; thus the same movements are performed now in one half 
which were before performed alternately in the two. Hence it is 
that a pipe with only one end closed, and a pipe of twice its length, 
with both ends closed, give the same pitch. 

329. Both Ends of the Pipe Open. When both ends of 
a pipe are open, it may still produce a musical tone, by having a 
node in the centre of it, thus forming two pipes like the one last 
described. When the vibration is established in such a pipe, the 
pulses from the ends move simultaneously toward C (Fig. 
and again from it after re 
flection. Thus C is a fixed *** 197 - 
point, where the greatest 
condensation and rarefac 

tion occur alternately, like ^= ~~c~ ^B 

A in Fig. 194. It there 
fore has the same pitch as A C alone, stopped at C and open at A. 
If a solid partition be inserted at (7, it causes no change of pitch. 
Such a pipe can produce no sound, except by the formation of 
at least one node. 

330. The Second Kind of Pipe is the Elementary 
Form. In comparing with each other tho three kinds of pipe 
which have been described, it is observable that the first kind 
(stopped at both ends), and the third kind (open at both ends), is 
each a double pipe of the second kind (open at one end, and 
stopped at the other). For, if two pipes of the second kind be 
placed with their open ends together, as we have seen, they form 
one of the first kind, and there is no change of pitch. Again, if 
the two be placed with the closed ends in contact, they form a pipe 
of the third class ; since the partition may remain or be removed, 
without affecting the mode of vibration. Hence, a pipe open at 
both ends, and one of the same length closed at both ends, each 

208 SOUND. 

yields the same fundamental note as a pipe of half their length, 
open only at one end. 

331. Vibrations of a Column of Air in Parts. The 
same is true of a column of air as of a string, that it may vihrate 
in parts ; and also, that two or more modes of vibration may co 
exist in the same column. 

The first and third kinds of pipe can divide so that the whole 
and the vibrating segments have the ratios of 1 : 4 : \ : &c.; these 
ratios in the closed pipe are shown in Figs. 194, 198, and 199 ; and 
in the open pipe in Figs. 197 and 200. In Fig. 198 the pipe is 
divided into two equal 
parts, in each of which Fia 

the vibrations take place 
in the same manner as in 
the whole, Fig. 194, 
Condensations run si 
multaneously from A and B to the middle point C, and thence 
back to A and B. When C is condensed, A and B are rarefied ; 
and when A and B are condensed, C is rarefied. Those three 
points have no amplitude, but the greatest changes in density. 
But the points midway between have the greatest amplitude, and 
no change of density. As the waves run over the parts in half 
the time that they would over the whole, the pitch is raised accord 
ingly. In this mode of vibrating, the opening where the vibra 
tions are excited cannot be at C, where the node is formed. 

In Fig. 199 are shown 

three vibrating segments. FlG - 199 - 

B and D are condensed at 
one moment, A and E at 

. JL JJ C JB -B 

In the third kind, as 

already stated (Art. 329), there must be at least one node. When 
there are two, it is apparent by Fig. 200 that they must be one- 
fourth of the length from 

each end, in order that the ^ FlG - 20 - 

three parts may vibrate in 

unison; for the middle 

part is a complete segment, ^r* " 

like the pipe A B (Fig. 

194), while the ends are half segments, like the pipe A C (Fig. 

196). If there were three nodes, there would be two complete 

segments between them, and two half segments at the ends. It is 

evident that the lengths of the half segments, being , |, , &c., 

are as 1, J, |, &c., of the whole pipe ; therefore the rates of vibra- 



FIG. 201 . 


tion (being inversely as the lengths) are as the numbers 1, 2, 3, 

In the second kind of pipe the ratios of length for successive 
modes of vibration are 1 : -J : -J, &c. The simplest division is by 
one node, a third of the length from the open end, as in Fig. 201. 
Then C D, a half segment, and A D, a com 
plete segment, have the same rate of vibra 
tion. If there were two nodes, one must 
be a fifth from the open end, while the 
other divides the remainder into two com- ^ 
plete segments. Therefore, in the several 

modes of vibration of the second kind of pipe, the half segments, 
being 1, I, , &c., of the whole length, the rates of vibration in 
them are as the odd numbers 1, 3, 5, &c. 

332. Modes of Exciting Vibrations in Pipes.. There 
are two methods of making the air-column in a pipe to vibrate: 
one by a stream of air blown across an orifice in the pipe, the 
other by an elastic plate called a reed. A familiar example of the 
first is the flute. A stream of air from the lips is directed across 
the embouchure, so as just to strike the opposite edge ; this causes 
a wave to move through the tube. The stream of air, like a spring, 
vibrates so as to keep time with the movement of the wave to and 
fro, while at each pulse it renews that movement, and makes the 
sound continuous. For higher notes, the stream must be blown 
more swiftly, that by its greater elastic force, it may be able to 
conform to the more rapid vibration of the column. A large pro 
portion of the pipes of an organ are made to produce musical 
tones essentially in the same way as the flute, and are called mouth- 

Fig. 202 shows the construction of the mouth-pipe of an organ ; 
o b is the mouth ; and as the stream of air issues from 
the channel i, it starts a wave in the pipe, and then 
the stream itself vibrates laterally past the lip b, keep 
ing time with the successive returns of the wave in 
the pipe. The pipe is attached to the wind-chest by 
the foot P. 

The clarinet is an example of vibrations in an air- 
column by a reed. In that instrument the reed is 
often made of wood; when the air is blown past its 
edge into the tube, the reed is thrown into vibration, 
and by it the column of air. The strength of elasticity 
in the reed should be such that its vibrations will keep 
time with the excursions of the wave in the column. 
What are called the reed pipes of the organ are con- 

FIG. 202. 



structed on the same principle, but the reeds are metallic. An ex 
ample is seen in Fig. 203, which represents a model of the reed 
pipe, made to show the vibrations through the 
glass walls at K A chimney, //, is usually FlG - 203< 

attached, sometimes of a form (as in the figure) 
to increase the loudness of the sound, and 
sometimes of a different form, for softening it. \ f 

333. Vibrations of Rods and Lam 
inae. A plate of metal called a reed is much 
used for musical purposes in connection with 
a column of air, as already stated. Except in 
such connection, the sounds of wires and lam 
inae are generally too feeble to be employed in 
music. But their vibrations have been much 
studied, on account of the interesting phe 
nomena attending them. 

334. Wires. If one end of a steel wire 
is fastened in a vise and vibrated, while a thin 
blade of lunlight falls across it, the path of the 
illuminated point may be traced. It is not 
ordinarily a circular arc about the fixed point 
as a centre, but some irregular figure; and 

frequently the point describes two systems of ellipses, the vibra 
tions passing alternately from one system to the other several 
times before running down. If the structure of the wire were the 
same in every part across its section, and if the fastening pressed 
equally on every point around it, the orbit of each particle would 
be a series of ellipses, whose major axes are on the same line. If, 
moreover, there was no obstruction to the motion, and the law of 
elasticity could obtain perfectly, it would vibrate in the same 
elliptic orbit forever, the force toward the centre being directly 
as the distance. It is easy to cause the wire, in the experiment 
just described, to vibrate also in parts ; in which case each atom, 
while describing the elliptic orbit, will perform several smaller 
circuits, which appear as waves on the circumference of the larger 

335. Chladni s Plates. If a square plate of glass or elastic 
metal, of uniform thickness and density, be fastened by its centre 
in a horizontal position, and a bow be drawn on its edge, it will 
emit a pure musical tone ; and by varying the action of the bow, 
and touching different points of the edge with the finger, a variety 
of sounds may be obtained from it. The plate necessarily vibrates 
in parts ; and the lowest pitch is produced when there are two 


nodal lines parallel to the sides, and crossing at the centre, thus 
dividing the plate into four square ventral segments. The posi 
tion of the nodal lines, and the forms of the segments, are beauti 
fully exhibited by sprinkling writing-sand on the plate. The par 
ticles will dance about rapidly till they find the lines of rest, where 
they will presently be collected. For every new tone the sand will 
show a new arrangement of nodal lines ; and as two or more 
modes of vibration may coexist in plates, as well as in strings and 
columns of air, the resultant nodes will also be rendered visible. 
Again, by fastening the plate at a different point, still other ar 
rangements will take place, each distinguishable by the position 
of its nodal lines and the pitch of its musical note. The form of 
the plate itself may also be varied, and each form will be charac 
terized by its own peculiar systems. Chladni, who first performed 
these interesting experiments, delineated and published the forms 
of ninety different systems of vibration in the square plate alone. 

If a fine light powder, as lycopodium (the pollen of a species 
of fern), be scattered on the plate, it is affected in a very different 
manner from heavy sand. It will gather into rounded heaps on 
those portions of the segments which have the greatest amplitude 
of vibration ; the particles which compose the heaps performing a 
continual circulation, down the sides of the heaps along the plate 
to the centre, and up the axis. If the vibration is violent, the 
heaps will be thrown up from the plate in little clouds over the 
portions of greatest motion. The cause of this singular effect was 
ascertained by Faraday, who found that in an exhausted receiver 
the phenomenon ceased. It is due to a circulation of the air, 
which lies in contact with a vibrating plate. The air next to 
those parts which have the greatest amplitude is at each vibration 
thrown upward more powerfully than elsewhere, and surrounding 
particles press into its place, and thus a circulation is established; 
and a fine light powder is more controlled by these atmospheric 
movements than by the direct action of the plate. 

336. Bells. If a thin plate of metal takes the form of a cyl 
inder or bell, its fundamental note is pro- F 
duced when each ring of the material 
changes from a circle to an ellipse, and 
then into a second ellipse, whose axis is 
at right angles to the former, as seen in 
Fig. 204. It thus has four ventral seg 
ments and four nodal lines, the latter ly 
ing in the plane of the axis of the bell or 
cylinder. If the rings which compose 
the bell were all detached from one an- 

212 SOUND. 

ether, they would have different rates of vibration according to 
their diameter, and hence would produce tones of various pitch ; 
but, being bound together by cohesion, they are compelled to keep 
the same time, and hence give but one fundamental tone. But a 
bell, especially if quite thin, may be made to emit a series of har 
monic sounds by dividing up into a greater number of segments. 
It is obvious that the number of nodes must always be even, be 
cause two successive segments must move in opposite directions 
in one and the same instant ; otherwise the point between them 
could not be kept at rest, and therefore would not be a node. Be 
sides the principal tone of a church-bell, one or two subordinate 
sounds on a different pitch may usually be detected. A glass bell, 
suitably mounted for the lecture-room, will yield ten or twelve 
harmonics, by means of a bow drawn on its edge. 

337. The Voice. The vocal organ is complex, consisting of 
a cavity called the larynx, and a pair of membranous folds like 
valves, having a narrow opening between them ; this opening, 
called the glottis, admits the air to the larynx from the wind-pipe 
below. The edges of these valves are thickened into a sort of 
cord, and for this reason the apparatus is called the vocal cords. In 
the act of breathing, the folds of the glottis lie relaxed and sepa 
rate from each other, and the air passes freely between them, with 
out producing vibration. But in the effort to form a vocal sound, 
they approach each other, and become tense, so that the current 
of air throws them into vibration. These vibrations are enforced 
by the consequent vibrations in the air of the larynx above ; and 
thus a fullness of sound is produced, as in many musical instru 
ments, in which a reed, and the air of a cavity, perform synchro 
nous vibrations, and emit a much louder sound than either could 
do alone. If two pieces of thin india-rubber be stretched across 
the end of a tube, with their edges parallel, and separated by a 
narrow space, as represented in Fig. 205, the ar 
rangement will give an idea of the larynx and 
glottis of the vocal organ. If air be forced through, 
a sound is produced, whose pitch depends on the 
size of the tube and the tension of the valves. 

The natural key of a person s voice depends on 
the length and weight of the vocal cords, and the 
size of the larynx. The yielding nature of all the 
parts, and the ability, by muscular action, to change 
the form and size of the cavity and the tension of the valves, give 
great variety to the pitch, and the power of adjusting it with pre 
cision to every shade of sound within certain limits. No instru 
ment of human contrivance can be brought into comparison with 



the organ of voice. After the voice is formed by its appropriate 
organ, it undergoes various modifications, by means of the palate, 
the tongue, the teeth, the lips, and the nose, before it is uttered in 
the form of articulate speech. 

338 The Organ of Hearing. The principal parts of the 
ear are the following : 

1. The outer ear, E a (Pig. 20G), terminating at the membrane 
of the tympanum, m. . 

FIG. 206. 

2. The tympanum, a cavity separated from the outer ear by a 
membrane, m, and containing a series of four very small bones 
(ossicles), b, c, o, and s, severally called, on account of their form, 
the hammer, the anvil, the ball, and the stirrup. The figure rep 
resents the walls of the tympanum as mostly removed, in order to 
show the internal parts. This cavity is connected with the back 
part of the mouth by the Eustachian tube, d. 

3. The labyrinth, consisting of the vestibule, v, the semicircular 
canals, f, and the cochlea, g. The latter is a spiral tube, winding 
two and a half times round. The parts of the labyrinth are exca 
vated in the hardest bone of the body. The figure shows only its 
exterior. There are two orifices through the bone which separates 
the labyrinth from the tympanum, the round orifice, e, passing 
into the cochlea, and the oval orifice, s, leading to the vestibule. 
These orifices are both closed by a thin membrane. The ossicles 
of the tympanum form a chain which connects the centre of the 
membrane, m, with that which closes the oval orifice. The laby 
rinth is filled with a liquid, in various parts of which float the 
fibres of the auditory nerve. 

By the form of the outer ear, the waves are concentrated upon 
the membrane of the tympanum, thence conveyed through the 

214 SOUND. 

chain of bones to the membrane of the labyrinth, and by that to 
the liquid within it, and thus to the auditory nerve, whose fibres 
lie in the liquid. 



339. Numerical Relations of the Notes. To obtain the 
series of notes which compose the common scale of music, it is 
convenient to use the mono chord. Calling the sound, which is 
given by the whole length of the string, the fundamental, or key 
note, of the scale, we measure off the following fractions of the 
whole for the successive notes, namely: f, {, f, ~, f, T 8 ,, 1. If the 
whole, and these fractions, are made to vibrate in order, the ear 
will recognize the sounds as forming the series called the gamut, 
or diatonic scale. And the interval between the fundamental and 
each of the others is named according to its distance inclusively. 
Thus, the interval from the whole (= 1) to f , is called the second; 
from 1 to |, the third, &c. ; therefore, from 1 to \, the eighth, or 
octave. Now, as the number of vibrations varies inversely as the 
length of the string, the numbers corresponding to the notes re 
spectively, are expressed by the same fractions inverted, 1, f , f , |, 
|, |, Jjf , 2. Reducing these to a common denominator, and using 
the numerators (since they have the same ratios), we have the fol 
lowing series, 24, 27, 30, 32, 36, 40, 45, 48, to express in the sim 
plest manner the relative numbers of vibrations in the notes of 
the scale, however produced. The sounds represented by these 
numbers are not arbitrarily chosen to form the scale, but they are 
demanded by the ear, and constitute the basis of the music of all 
ages and nations. 

340. Interval of the Second. In examining the relation 
of each two successive numbers in the foregoing series, we find 
three different ratios. Thus, 

27 : 24, 36 : 32, and 45 : 40, is each as 9 : 8. 

30 : 27 and 40 : 36, 10 : 9. 

32 : 30 and 48 : 45, 16 : 15. 

Therefore, of the seven intervals, called the second, in the dia 
tonic scale, there are three equal to f , two equal *, and two others 
equal to |f . Each of the first five is called a major second, or a 
tone ; each of the last two is called a minor second, or a semitone. 


One of the larger tones exceeds one of the smaller by g J ; for 
I -r Jy = JJ. This small interval, gj, is called a comma, and is 
employed as a measuring unit in estimating the relations of inter 
vals. The minor second, though called a semitone, is in fact 
more than half of either kind of major second. 

341. Repetition of the Scale. The eighth note of the 
scale so much resemhles the first in sound, that it is regarded as a 
repetition of it, and called by the same name. Beginning, there 
fore, with the half string, where the former series closed, let us 
consider the sound of that as the fundamental, and take of it 
for the second, f of it for the third, &c. ; we then close a second 
series of notes on the quarter-string, whose sound is also consid 
ered a repetition of the former fundamental. Each fraction of 
the string used in the second scale is obviously half of the corre 
sponding fraction of the whole string, and therefore its note an 
octave above the note of that. This process may be repeated in 
definitely, giving the second octave, third octave, &c. Ten or 
eleven octaves comprehend all sounds appreciable by the human 
ear ; the vibrations of the extreme notes of this entire range have 
the ratio of 1 : 2 , or 1:2"; that is, 1 : 1024, or 1 : 2048. Hence, 
if 16 vibrations per second produce the lowest appreciable note, 
the highest varies from 16,000 to 33,000. It was ascertained by 
Dr. Wollaston that the highest limit is different for different ears ; 
so that when one person complains of the piercing shrillness of a 
sound, another maintains that there is no sound at all. The 
lowest limit is indefinite for a different reason ; the sounds are 
heard by all, but some will recognize them as low musical tones, 
while others only perceive a rattling or fluttering noise. Few mu 
sical instruments comprehend more than six octaves, and the 
human voice has only from one to three, the male voice being in 
pitch an octave lower than the female. 

342. Modes of Naming the Notes. There is one system 
of names for the notes of the scale, which is fixed, and another 
which is movable. The first is by the seven letters, A,B, C, D, E, 
F, G. The notes of the second octave are expressed by the same 
letters, in some way distinguished from the former. The best 
method is to write by the side of the letter the numeral expressing 
that index of 2, which corresponds to the octave: as A^ A^ &c., 
in the octaves above; A,, A^ in those below. 

The second mode of designation is by the syllables, do, re, mi, 
fa, sol, la, si. These express merely the relations of notes to each 
other, do always being the fundamental, re its second, mi its third, 
&c. In the natural scale, do is on the letter C, re on D, &c. ; but 
by the aid of interpolated notes, the scale of syllables may be 

216 SOUND. 

transferred, so as to begin successively with every letter of the 
fixed scale. 

343. The Chromatic Scale. Let the notes of the diatonic 
scale be represented (Fig. 207) by the horizontal lines, C, D, &c.; 
the distance from C to D being a tone, from D to E a 

tone, E to F a semitone, &c. It will be observed that IQ 
the fundamental, C, is so situated that there are two - 
whole tones above it, before a semitone occurs, and then 
three whole tones before the next semitone. C is there- 
fore the letter to be called by the syllable do, in order to 
bring the first semitone between the 3d and 4th, and 
the other semitone between the 7th and 8th, as the 
figure represents them. Now, that we may be able to 
transfer the scale of relations to every part of the fixed 
scale (which is necessary, in order to vary the character 
of music, without throwing it beyond the reach of the 
voice), the whole tones are bisected, and two semitone 
intervals occupy the place of each. The dotted lines in 
the figure show the places of the interpolated notes, 
which, with the original notes of the diatonic scale, di- - 
vide the whole into a series of semitones. This is called 
the chromatic scale. The interpolated note between G 
and D is written C$ (C sharp), or Dfc (D flat), and so 
of the others. As the whole tones lie in groups of twos 
and threes, so the new notes inserted are grouped in the 
same way. This explains the arrangement of the black 
keys by twos and threes alternately in the key-board of the organ 
and piano-forte. The white keys compose the diatonic scale, the 
white and black keys together, the chromatic scale. It is obvious 
that on the chromatic scale any one of the twelve notes which 
compose it may become do, or the fundamental note, since the re 
quired series, 2 tones, 1 semitone, 3 tones, 1 semitone, can be ar 
ranged to succeed each other, at whatever note we begin the 
reckoning. This change, by which the fundamental note is made 
to fall on different letters, is called the transposition of the scale. 

344. Chords and Discords. When two or more sounds, 
meeting the ear at once, form a combination which is agreeable, 
it is called a chord ; if disagreeable, a discord. The disagreeable 
quality of a discord, if attended to, will be perceived to consist in 
a certain roughness, or harshness, however smooth and pure the 
simple sounds which are combined. On examining the combina 
tions, it will be found that if the vibrations of two sounds are in 
some very simple relations, as 1 : 2, 1 : 3, 2 : 3, 3 : 4, &c., they pro 
duce a chord ; and the lower the terms of the ratio, the more per- 


fret the chord. On the other hand, if the numbers necessary to 
i-x press the relations of the sounds are large, as 8 : 9, or 15 : 16, a 
discord is produced. It appears that concordant sounds have/re- 
qwni coincidences of vibrations. If, in two sounds, there is coin 
cidence at every vibration of each, then the pitch is the same, and 
the combination is called unison. If every vibration of one coin 
cides with every alternate vibration of the other, the ratio is 1 : 2, 
and the chord is the octave, the most perfect possible. Tfaejifth is 
the next most perfect chord, where every second vibration of the 
lower meets every third of the higher, 2 : 3. The fourth, 3 : 4, the 
major third, 4 : 5, the minor third, 5 : G, and the sixth, 3 : 5, are 
reckoned among chords; while the second, 8 : 9, and the seventh, 
8 : 15, are harsh discords. What is called the common chord con 
sists of the 1st, 3d, and 5th, combined, and is far more used in 
music than any other. Harmony consists of a succession of chords, 
or rather, of such a succession of combined sounds as is pleasing 
to the ear; for discords are employed in musical composition, 
their use being limited by special rules. Many combinations, 
which would be too disagreeable for the ear to dwell upon, or to 
finish a musical period, are yet quite necessary to produce the best 
effect ; and without the relief which they give, perfect harmony, 
if long continued, would satiate. 

345. Temperament. This is a term applied to the small 
eiTors introduced into the notes, in tuning an instrument of fixed 
keys, in order to adapt the notes equally to the several scales. If 
the tones were all equal, and if semitones were truly half tones, no 
such adjustment of notes would be needed; they would all be ex 
actly correct for every scale. Representing the notes in the scale 
whose fundamental is C by the numbers in Art. 339, we have, 

C, D, E, F, G, A, B, C,, D., 9 E., &c. 

24, 27, 30, 32, 36, 40, 45, 48, 54, 60, &c. 

Now suppose we wish to make D, instead of C, our key-note ; 
then it is obvious that E will not be exactly correct for the second 
on the new scale. For the fundamental to its second is as 8 to 9 ; 
and 8 : 9 : : 27 : 30.375, instead of 30. Therefore, if D is the key 
note, we must have a new E, slightly above the E of the original 
scale. So we find that A, represented by 40, will not serve to be 
the 5th in the new scale ; since 2 : 3 : : 27 : 40.5, which is a little 
higher than A ( 40). After adding these and other new notes, 
to render the intervals all exactly right for the new key of D, if 
vve proceed in the same manner, and make E (= 30) our key-note, 
and obtain its second, third, &c., exactly, we shall find some of 
them differing a little, both from those of the key of C, and also 

318 SOUND. 

cf the key of D. Using in this way all the twelve notes of the 
chromatic scale in succession for the fundamental, it appears that 
several different E s, F s, G s, &c., are required, in order to make 
each scale perfect. In instruments, whose sounds cannot be mod 
ified by the performer, like the organ and piano-forte, as it is con 
sidered impossible to insert all the pipes or strings necessary to 
render every scale perfect, such an adjustment is made as to dis 
tribute these errors equally among all the scales. For example, E 
is not made a perfect third for the key of #, lest it should be too 
imperfect for a second in the key of D, and for its appropriate 
place in other scales. It is this equal distribution of errors among 
the several scales which is called temperament. The errors, when 
thus distributed, are too small to be observed by most persons ; 
whereas, if an instrument was tuned perfectly for any one scale, 
all others would be intolerable. 

The word temperament, as above explained, has no application 
except to instruments of fixed keys, as the organ and piano-forte ; 
for, where the performer can control and modify the notes as he is 
playing, he can make every key perfect, and then there are no 
errors to be distributed. The flute-player can roll the flute 
slightly, and thus humor the sound, so as to cause the same 
fingering to give a precisely correct second for one scale, a correct 
third for another, and so on. The player on the violin does the 
same, by touching the string in points slightly different. The or 
gans of the voice, especially, can be adjusted to make the intervals 
perfect on every scale. In these cases there is no tempering, or 
dividing of errors among different scales, but a perfect adjustment 
to each scale, by which all error is avoided. 

346. Harmonics. The fact has been mentioned that a string, 
or a column of air, may vibrate in parts, even while vibrating as a 
whole. It only remains to show the musical relations of the 
sounds thus produced. When a string vibrates in parts, it divides 
into halves, thirds, fourths, or other aliquot parts. Now, a half- 
string produces an octave above the whole, making the most per 
fect chord with it. The third of a string being two-thirds of the 
half-string, produces the fifth above the octave, a very perfect 
chord. The quarter-string gives the second octave ; the fifth part 
of it, being | of the quarter, gives the major third above the 
second octave ; and the sixth part, being f of the quarter, gives 
the fifth above the second octave. Thus, all the simpler divisions, 
which are the ones most likely to occur, are such as produce the 
best chords ; and it is for this reason that the sounds are called 
harmonics. The same is true of air-columns and bells. The 
JSolian harp furnishes a beautiful example of the harmonics of a 


string. Two or more fine smooth cords are fastened upon a box, 
and tuned, at suitable intervals, like the strings of a violin ; and 
the box is placed in a narrow opening, where a current of air 
passes. Each string at different times, according to the intensity 
of the breeze, will emit a pure musical note ; and, with every 
change, will divide itself in a new mode, and give another pitch, 
while it will frequently happen that the vibrations of different 
divisions will coexist, and their harmonic sounds mingle with each 

347. Overtones. But the parts into which a sounding body 
divides do not always harmonize with the whole. For instance, J 
or T T of a string is discordant with the fundamental. The word 
harmonics is not, therefore, applicable except to a very few of the 
many possible sounds which a body may produce. The word 
overtone is used to express in general any sound whatever, given 
by a part of a sounding body. A string may furnish 20 or 30 
overtones, but only a small number of them would be harmonics. 

348. Quality of Tone. Even when the pitch of two sound 
ing bodies is the same, the ear almost always distinguishes one 
sound from the other by certain qualities of tone peculiar to each. 
Thus, if the same letter be sounded by a flute and the string of a 
piano, each note is easily distinguished from the other. Two 
church-bells may be upon the same key, and yet one be agreeable, 
and the other harsh to the ear. While these great diversities may 
to some extent be due to circumstances not yet discovered, still it 
is certain that they in no small degree arise from the vibrations 
of various parts mingling with those of the fundamental sound. 
A long monochord can, by varying the mode of exciting the vibra 
tions, be made to yield a great variety of sounds, while there is 
perceived in them all the same fundamental undertone which de 
termines the pitch. If the string be struck at the middle, then no 
node can be formed at that point; hence, the mixed sound will 
contain no overtones of the J, \, , |, T , or other even aliquot 
parts of the string ; for all such would require a node at the mid 
dle. But if struck at one-third of its length from the end, then 
the overtones, -J, -}, &c., may exist, but not those of J, 1, /,, or any 
other parts whose node would fall at \ of the length from the 

For reasons which are mostly unknown, some sounding bodies 
have their fundamental accompanied by harmonic overtones, and 
others by overtones which are discordant. And this is one cause 
of the agreeable or unpleasant quality of the sounds of different 

220 SOUND. 

349. Communication of Vibrations. The acoustic vibra 
tions of one body are readily communicated to others, which are 
near or in contact. "We have already noticed that the vibrations 
of a reed will excite those of a column of air in a pipe. If two 
strings, which are adapted to vibrate alike, are fastened on the 
same box, and one of them is made to sound, the other will sound 
also more or less loudly, according to the intimacy of their connec 
tion. The vibrations are communicated partly through the air, 
and partly through the materials of the box. So, if a loud sound 
is uttered near a piano-forte, several strings will be thrown into 
vibration, whose notes are heard after the voice ceases. The no 
ticeable fact in all sdch experiments is, that the vibrations thus 
communicated from one body to another cause sounds which har 
monize Avith each other, and with the original sound. For the 
rate of vibration will either be identical, or have those simple re 
lations which are expressed by the smallest numbers. Let a per 
son hold a pneumatic receiver or a large tumbler before him, and 
utter at the mouth of it several sounds of different pitch ; and he 
will probably find some one pitch which will be distinctly rein 
forced by the vessel. That particular note, which the receiver by 
its size and form is adapted to produce, will not be called forth by 
a sound that would be discordant with it. The melodeon, ser- 
aphine, and instruments of like character, owe their full and bril 
liant notes to reeds, each of which has its cavity of air adapted to 
vibrate in unison with it. It sometimes happens that the second 
body, vibrating as a whole, would not harmonize with the first, and 
yet will give the same note by some mode of division. Thus it is 
that all the various sounds of the monochord, and of the strings 
of the viol, are reinforced by the case of thin wood upon which 
they are stretched. The plates of wood divide by nodal lines into 
some new arrangement of ventral segments for every new sound 
emitted by the string. In like manner, the pitch of the tuning- 
fork, and all the rapid notes of a music-box, are rendered loud and 
full by the table, in contact with which they are brought. The 
extended material of the table is capable of division into a great 
variety of forms, and will always give a sound in unison with the 
instrument which touches it. 

350. One System of Vibrations Controlling Another. 

If two sounding bodies are nearly, but not precisely on the same 
key, they will sometimes, when brought into close contact, be 
made to harmonize perfectly. The vibrations of the more power 
ful will be communicated to the other, and control its movements 
so tha t the discordance, which they produce when a few inches 
apart, will cease, and concord will ensue. Two diapason pipes of 


an organ, timed a quarter-tone or even a semitone from unison, so 
as to jar disagreeably upon the ear, when one inch or more asun 
der, will be in perfect unison, if they are in contact through their 
whole length. Even the slow oscillations of two watches will in 
fluence each other ; if one gains on the other only a few beats in 
an hour, then, if they are placed side by "side on the same board, 
they will beat precisely together. 

351. Crispations of Fluids. Among the numerous acous 
tic experiments illustrating the communication of vibrations, none 
are more beautiful than those in which the vibrations of glass rods 
are conveyed to the surface of a fluid. Let a very shallow pan of 
glass or metal be attached to the middle of a thin bar of wood, 
three or four feet long, and resting near its ends on two fixed 
bridges ; let water be placed in the pan, and a long glass rod 
standing in it, or on the wood, be vibrated longitudinally, by 
drawing the moistened fingers down upon it ; the liquid immedi 
ately shows that the vibrations are communicated to it. The sur 
face is covered with a regular arrangement of heaps, called crispa- 
tions, which vary in size with the pitch of sound, which is produced 
by the same vibration. If the pitch is higher, they are smaller, 
and may be readily varied from three or four inches in diameter to 
the fineness of the teeth of a file. Crispations of the same charac 
ter are also formed in clusters on the water in a large tumbler or 
glass receiver, when the finger is drawn along its edge ; every ven 
tral segment of the glass produces a group of hillocks by the side 
of it on the surface of the water. 

352. Interference of "Waves of Sound. Whenever two 
sounds are moving through the air, every particle will, at a given 
instant, have a motion which is the resultant of the two motions 
which it would have had if the sounds were separate. These mo 
tions may conspire, or they may oppose each other. The word in 
terference is used in scientific language to express the resultant 
effect, whatever it may be. The beats, which are frequently heard 
in listening to two sounds, indicate the points of maximum con 
densation produced by the union of the condensed parts of both 
systems of waves. And the sounds are considered discordant when 
these beats are just so frequent as to produce a disagreeable flut 
tering or rattling. If too near or too far apart for this, they are 
regarded practically as concordant. And when the beats are too 
close to be perceived separately, yet the peculiar adjustment of 
condensations of one system with those of the other, according as 
one wave measures two, or two waves measure three, or four 
measure five, &c., is at once distinguished by the ear, and recog 
nized as the chord of the octave, the fifth, the third, &c. When a 

222 SOUND. 

sound and its octave are advancing together, there are instants in 
which any given particle of air is impressed with two opposite mo 
tions, and other alternate moments when both motions are iri the 
same direction. For the waves of the highest sound are half as 
long as those of the lowest ; hence, while every second condensa 
tion of the former coincides with every condensation of the latter, 
the alternate ones of the former must be at the points of greatest 
rarefaction of the latter ; and this cannot occur without opposite 
movements of the particles. If two simultaneous sounds have the 
same pitch, i. e., the same length of wave, they ordinarily run to 
gether, so that like phases in the two systems are coincident, and 
the compound sound (called unison] simply has twice the loiidness 
of one of them alone. But, by a delicate mode of experiment, one 
of these two systems of waves, having equal lengths, and equal in 
tensities, may be made to fall half a wave behind the other, in 
which case opposite phases coincide, and the two sounds destroy 
each other. Thus, two sounds produce silence) on the same prin 
ciple that two systems of water-waves may produce level water. 

353. Number and Length of Waves for Each Note. 

Though the vibrations of any musical note are too rapid to be 
counted, yet the number may be ascertained in several ways. One 
of the readiest methods is by means of a little instrument called 
the siren, invented by De La Tour. The pulses are produced by 
streams of air driven through holes in a revolving wheel. The 
revolutions of the wheel are recorded by machinery, and the num 
ber of vibrations in each revolution is known from the number of 
holes through which the air rushes. "When such a velocity of 
revolution is given as to produce the required pitch, then the rev 
olutions of the index per minute may be counted, and the number 
of vibrations in the same time will be known, and therefore the 
number per second. In this and other ways it is ascertained that 
the numbers corresponding to the letters of the scale are the 
following : 

( C, A E, F, G, A, B, C 2 , D,, 
(128, 144, 160, 170f, 192, 213|, 240, 256, 288. 

The highest note of the above series, D. 2 , 288, is the lowest on the 
common or /Xflute. There is not, however, a perfect agreement 
of pitch in different countries, and among different classes of mu 
sicians. Accordingly, (7, which is given above as corresponding to 
128 vibrations per second, has several values, varying from 127 to 

To find the length of acoustic waves for any given pitch, we 
have only to divide the velocity of sound in one second by the 


number of vibrations which roach the ear in the same length of 
time. For example, at the temperature of 60, sound travels 1118 
feet per second ; therefore the length of waves of low D on the 
flute = 1118 -r- 288 = nearly four feet. The waves of the lowest 
musical note are about 70 feet long ; and of the highest, less than 
half an inch. 

354. Acoustic Vibrations Visibly Projected. The vi 
brations of heavy tuning-forks can be magnified and rendered dis 
tinctly visible to an audience by projecting them on a screen. 
The fork being constructed with a small metallic mirror attached 
near the end of one prong, a sunbeam reflected from the mirror 
will exhibit all the movements of the fork greatly enlarged on a 
distant wall ; and if the fork is turned on its axis, the luminous 
projection will take the form of a waving line. And by the use 
of two forks, all the phenomena of interference may be rendered 
as distinct to the eye as they are to the ear. 




355. The Magnet. Fragments of iron ore are sometimes 
found which strongly attract iron ; and bars of steel are artificially 
prepared which exhibit the same property. These bodies are 
called magnets ; the ore is the natural magnet, commonly called 
lodestone; the prepared steel bar is an artificial magnet. 

Another property distinguishes the magnet, namely, that when 
properly suspended on a pivot, it assumes a certain definite direc 
tion with regard to the earth. This property of the magnet is 
called its polarity. 

355. The Attraction Between a Magnet and Iron. 

The magnetic property which is likely to be first noticed is the 
attraction of iron. If a lodestone or a bar magnet be rolled in 
iron filings (Fig. 208), there are two opposite points to which the 

FIG. 208. 

filings attach themselves in thick clusters, arranged in diverging 
filaments. These opposite points of greatest action are called 
poles. The attraction diminishes from the poles towards the cen 
tral parts ; and about in the middle between them there is little 
or none ; this is called the neutral point. The straight line from 
one pole to the other is called the axis. 

The mutual attraction between a magnet and iron is shown 
by bringing a piece of iron toward either pole of the magnetic 
needle ; the needle instantly turns so as to bring its pole as near 



FIG. 209. 

as possible to the iron (Fig. 209). On the other hand, an iron 
needle being suspended in 
like manner, the same move 
ment takes place, when 
either pole of a magnet is 
brought near to it 

N ickel, cobalt, and some 
times manganese, exhibit 
the same magnetic proper 
ties in some degree. But 
these exist in comparatively 
small quantities, and therefore by magnetic bodies are usually in 
tended only iron and steel. 

FIG. 210. 

357. Polarity. If a light magnet is delicately suspended on 
a pivot at the neutral point, as in Fig. 210, it is called a magnetic 
needle. When thus placed and left to it 
self, it oscillates for a time, and finally 

settles with its axis in a certain fixed di 
rection, which in most places is nearly 
north and south. The end which points 
in a northerly direction is called the north 
pole; the other, the south pole. These 
poles are usually marked on the larger 
magnets by the letters N and S, so that they may be instantly 
distinguished. If a magnetic needle has simply a mark or stain 
on one end, that end is understood to be the north pole. 

358. Action of Magnets on Each Other. While either 
pole of a magnet attracts and is attracted by a piece of iron, it is 
otherwise when the pole of one magnet is brought near the pole 
of another. There is attraction in some cases, and repulsion in 
others. If the magnets are properly marked, and one of them 
suspended so as to move freely, it is readily discovered that tiio 
law of action is the following : 

Poles of the same name repel, and 
those of contrary name attract each 

Thus, the pole 8 of the magnet 
(Fig. 211) repels s of the needle, and 
attracts n ; and if the magnet were 
inverted, and the pole N brought 
near to n, the latter would be re 
pelled, and s be attracted. 

FIG. 211. 


359. Magnetic Induction. When a bar of iron is brought 
near to the pole of a magnet, though attraction is the phenome 
non first observed, as stated (Art. 356), yet it is readily proved 
that this attraction results from a change which is previously pro 
duced in the iron. It becomes a magnet through the influence of 
the magnet which is near it. That end of the iron bar which is 
placed near one pole of a magnet becomes a pole of the opposite 
name, and the remote end a pole of the same name. Hence, ac 
cording to the law (Art. 358), the poles which are contiguous at 
tract each other, because they are unlike. The influence by which 
the iron becomes a magnet is called induction. A magnet, when 
brought near to a piece of iron, induces upon the iron the mag 
netic condition, without any loss of its own magnetic properties. 
This influence is more powerful according as the two are nearer to 
each other ; it is, therefore, greatest when the two bars are in 

That the iron is truly a magnet for the time being is proved 
by bringing a needle near to its remote end ; one pole is attracted, 
and the other repelled. If the iron had not been changed into a 
magnet, each pole of the needle would be attracted by it (Art. 356). 

360. Successive Inductions, Let a bar of iron, B (Fig. 
212), be suspended from the south pole of the magnet A ; then 
the upper end of B is a north pole, and the lower end a south pole. 
Now, as B is a magnet, it will in 
duce the magnetic state on another 

bar, (7, when brought in contact ; 
and, as before, the poles of opposite 
name will be contiguous. There 
fore, the upper end of C is north, 
arid the lower end south. D is also 
a magnet by the inductive power 
of C. Thus, there is an indefinite series of inductions, growing 
weaker, however, from one to another, as the number is greater, 
and as the bars are longer. 

The filaments of iron filings which attach themselves to the 
pole of a magnet (Art. 356) are so many series of small magnets 
formed in the same manner as just described. Every particle of 
iron is a complete magnet, having its poles so arranged that the 
opposite poles of two successive particles are always contiguous. 

361. Reflex Influence. When a magnet exerts the induc 
tive power upon a piece of iron which is near it, its own magnetic 
intensity is increased. The end of the piece of iron contiguous to 
the pole of the magnet is no sooner endued with the opposite polar 
ity than it reacts upon the magnet, and increases its intensity ; so 


that, if fragments of iron are attached to a magnet, as many as it 
will sustain, then after a time another may be added, and again 
another, till there is a very sensible increase of its original 

Hence, too, the force of attraction of the dissimilar poles of 
two magnets is greater than the force of repulsion of the similar 
poles; because, when the poles are unlike, each acts inductively 
on the other to develop its poles more fully ; but when they are 
alike, the influence which they reciprocally exert tends to make 
them unlike, and of course to diminish their repulsive force. 

An extreme case of this diminution of repulsive force occurs 
when the like poles of two very unequal magnets are brought into 
contact. The small magnet immediately clings to the large one, 
as though the poles were unlike ; and if examined, it is found that 
they are unlike. The powerful magnet has in an instant reversed 
the poles of the weak one by its strong inductive power, the latter 
not having force enough to diminish sensibly the strength of the 

362. Double Induction. The effects of two inductions at 
once on a bar of iron are various. 

1. The bar may become a single magnet of double strength. 

2. It may consist of two distinct magnets. 

3. It may have no magnetic power at all. 

The first case is illustrated by bringing the north pole of a 
magnet to one end of the iron, and the south pole of another mag 
net to the other end. Each magnet will form two poles by induc 
tion, and it is evident that the two pairs of poles will coincide. 
Even one magnet produces the same effect when laid by the side 
of a bar of iron of the same length. 

To show the second effect, apply one pole of a magnet to the 
middle of the iron bar ; then an opposite pole is formed at the 
middle, and a like pole at each end, each half of the bar being a 
separate magnet. The same effect is produced by bringing the 
like poles of two magnets in contact with the ends of the bar ; for 
both ends will be of the opposite kind, and the middle of the same 
kind, as the poles applied. If a pole is applied to the middle of a 
star of iron, the extremity of each ray is a pole of the same kind ; 
if to the middle of a circle of iron, the same polarity is found at 
every point of the circumference. 

As an example of the third case, suspend a bar of iron from 
the pole of a magnet, and then bring the opposite pole of an equal 
magnet to the point of contact ; the two poles induced by one are 
contrary to the two induced by the other, and they are found to 
be completely neutralized. 


This last case shows that two opposite and equal magnetic 
poles formed at the same point destroy each other. 

363. Coercive Force. If in the several experiments on iron 
bars, which have been already described, pure annealed iron is 
used, the effects take place instantly ; and when the magnet is re 
moved, they as suddenly disappear. But if the iron is hard, mag 
netic poles are developed in it slowly ; and when they have been 
developed, the iron returns also slowly to its neutral condition. 

That property of hard unannealed iron which obstructs the 
development of magnetism in it, and which hinders its return to 
a neutral state, is called the coercive force. In iron which is pure 
and well annealed there seems to be no coercive force. It appears 
in a slight degree in iron not carefully prepared, and increases 
with its hardness ; it is great in tempered steel, which is a com 
pound of iron and carbon, and greatest of all in steel, which is 
tempered to the utmost hardness. 

It is, therefore, difficult to make a strong magnet of a steel bar 
by ordinary induction, unless it is quite thin ; but after the de 
velopment has once been made, the bar becomes a permanent mag 
net, and may by care be used as such for years. 

364. Change in the Coercive Force. The coercive force 
is weakened by any cause which excites a tremulous or vibratory 
motion among the particles of the steel. This happens when the 
bar is struck by a hammer, so as to produce a ringing sound, 
which indicates that the particles are thrown into a vibratory mo 
tion. The passage of an electric discharge through a steel bar 
under the influence of a magnet, overcomes the coercive force for 
the time being, and permanent magnetism is developed. Heat 
produces the same effect ; and hence a steel bar is conveniently 
magnetized by heating it to redness, placing it under a powerful 
inductive influence, and then hardening it by sudden cooling. 
The coercive force is thus neutralized by heat, till the develop 
ment takes place, when it is restored, and the bar is a permanent 

A magnet, however, loses its power by the same means as, dur 
ing the process of induction, were used to develop it. Accord 
ingly, any mechanical concussion or rough usage impairs or 
destroys the power of a magnet. By falling on a hard floor, or 
by being struck with a hammer, it is injured. Heat produces a 
similar effect. A boiling heat weakens, and a red heat totally de 
stroys the magnetism of a needle. 

365. Magnetism not Transferred, but only Developed. 

This is strikingly proved by the fact that if a magnet be divided. 


even at the neutral point, where there is no sign of magnetism, 
the parts instantly become complete magnets, two unlike poles 
manifesting themselves at the place of fracture. Both polarities 
seem to exist at every point, and are developed wherever the bar is 
divided. If each part is divided again, the same phenomenon is 
repeated, and so on indefinitely. There is, therefore, no transfer 
of magnetism from one point to another, any more than from one 
bar to another, but only an excitation of what existed in every 
part of the body before. Both the north and the south pole must 
be conceived as latent at every point of a piece of iron or steel ; 
and when the piece is magnetized, either north or south polarity 
is developed more or less fully in all parts except the neutral point. 
It is not necessary that the particles should be united by cohe 
sion in a solid bar. A magnet can be formed by filling a brass 
tube with iron filings and sand, or by forming a rod of cement 
mixed with filings, and then subjecting them to inductive influ 
ence. Fig. 213 will give an 
idea of the probable struc- FlG - 2 

ture of every magnet. Each 
particle of it is a complete 
magnet, the like poles of all 

are turned the same way, and unlike poles are therefore contigu 
ous to each other, and each acts inductively on the next. 

366. Magnetic Intensity and Distance. The law of the 

magnetic force is the following : 

The intensity of the magnetic force, whether attraction or re 
pulsion, varies inversely as the square of the distance. 

The law in the case of the repulsion of like poles is readily 
proved by Coulomb s torsion balance, which is figured and de 
scribed in Art. 402, under Electricity. The angle of torsion is 
used as a measure of the repulsion, and it is found that the wire 
must be twisted through four times as large an angle to bring the 
poles to one-half the distance, and nine times as large an angle to 
bring them to one-third the distance, &c., the force increasing as 
the square of the distance diminishes. 

To prove the law for the attraction of opposite poles, the vibra 
tions of a needle are counted, when it is placed at different dis 
tances from a magnet. The square of the number made in a given 
time is a measure of the attractive force, just as the square of the 
number of vibrations of a pendulum is a measure of the force of 
gravity (Art. 170). 

In each of these experiments, the magnetic influence of the 
earth upon the needle must be eliminated, in order to obtain a 
correct result 



367. Equilibrium of a Needle Near a Maasiet. If a 

small needle, free to revolve, be placed near the pole of a magnet, 
so that its centre is in the axis of the magnet produced, it will 
place itself in the line of that axis. For suppose that N S (Fig. 
214) is a large magnetic 
bar, and n s a small needle 
suspended near the north 
pole of the magnet, with 
its centre in the axis of 
the bar produced at a ; it 
will be seen that the ac 
tion of the pole of the 
magnet is such as to bring 
the needle into a line with 
the magnet. The action of the pole N upon the needle tending 
to give it this direction (since it repels n and attracts s), is equal 
to the sum of its actions upon both poles. The pole S, by repel 
ling s, and attracting n, tends to reverse this position, but, on ac 
count of greater distance, its force is less than that of N. 

If the centre of the needle is in a line perpendicular to the bar 
at its middle point, the needle will be in equilibrium when paral 
lel to the bar with its poles in contrary order. Thus, supposing 
the needle to be suspended at b, it will be seen that the actions of 
both poles of the magnet conspire to move n to the left, and s to 
the right; and as these forces are equal, equilibrium takes place 
only when the needle is parallel to the bar. 

At intermediate points the needle will assume all possible in 
clinations to the axis of the bar, each position being determined 
by the resultant of the four forces which act on the needle. In 
Fig. 215 are indicated some of the positions which the needle takes 

FIG. 215. 

in being carried round the magnet. While it goes once round the 
magnet, it makes two revolutions on its own axis. 

It is to be observed that in all positions the needle tends, as a 



whole, to move toward the bar, since the attractions always exceed 
the repulsions. 

368. Magnetic Curves. All the foregoing cases are shown 
at once by iron filings strewn on paper or parchment, which is 
stretched on a frame and placed near a magnet. Let the paper be 
slightly jarred, while the magnet lies parallel to it, either above or 
below, and all the inclinations of the needle will be represented 
by the particles of iron arranged in curves from pole to pole (Fig. 
216). Near the poles of the magnet the filings stand up on the 

FIG. 216. 


paper at various inclinations. These are the extremities of still 
other curves, which would be formed in all possible planes passing 
through the axis of the magnet, provided the filings could float 
suspended in the air, while the magnet is placed in the midst of 
them. These are called magnetic curves. 

When the magnet is below the paper, the particles move away 
from the area over the poles, as in Fig. 216 ; but when it is above, 
they gather in a cluster under each pole. This singular difference 
arises from the force of gravity acting on the filaments, which are 
raised up on the paper, and which lean, in the former case, from 
each other, and in the latter, toward each other. 



369. Declination of the Needle. When the needle is bal 
anced horizontally, and free to revolve, it does not generally point 
exactly north and south ; and the angle by which it deviates from 
the meridian is called the declination. A vertical circle coinci- 



dent with the direction of the needle at any place is called the 
magnetic meridian. As the angle between the magnetic and the 
geographical meridians is generally different for different places, 
and also varies at different times in the same place, the word vari 
ation expresses these changes in declination, though it is much 
used as synonymous with declination itself. 

370. Isogonic Curves. This name is given to a system of 
lines imagined to be drawn through all the points of equal decli 
nation on the earth s surface. We naturally take as the standard 
line of the system that which connects the points of no declina 
tion, or the isogonic of (Fig. 217). Commencing at the north 

FIG. 217. 

pole of dip, about Lat. 70, Lon. 96, it runs in a general direc 
tion E. of S., through Hudson s Bay, across Lake Erie, and the 
State of Pennsylvania, and enters the Atlantic Ocean on the coast 
of North Carolina, Thence it passes east of the West India 
Islands, and across the N. E. part of South America, pursuing its 
course to the south polar regions. It reappears in the eastern hem 
isphere, crosses Western Australia, and bears rapidly westward 
across the Indian Ocean, and then pursues a northerly course 
across the Caspian Sea to the Arctic Ocean. There is also a de 
tached line of no declination, lying in eastern Asia and the Pacific 
Ocean, returning into itself, and inclosing an oval area of 40 N". 
and S. by 30 E. and. W. Between the two main lines of no decli 
nation in the Atlantic hemisphere, the declination is westward, 
marked by continued lines, in Fig. 217 ; in the Pacific hemisphere, 
outside of the oval line just described, it is eastward, marked by 
dotted lines. Hence, on the American continent, in all places 
east of the isogonic of 0, the north pole of the needle declines 
westward, and in all places west of it, the north pole declines east- 


ward ; on the other continent this is reversed, as shown by the 

Among other irregularities in the isogonic system, there are 
two instances in which a curve makes a wide sweep, and then in 
tersects its own path, while those within the loop thus formed re 
turn into themselves. One of these is the isogonic of 8 40 E., 
which intersects in the Pacific Ocean west of Central America ; 
the other is that of 22 13 W., intersecting in Africa. 

In the northeastern part of the United States the declination 
has long been a few degrees to the west, with very slow and some 
what irregular variations. 

371. Secular and Annual Variation. The declination of 
the needle at a given place is not constant, but is subject to a 
slow change, which carries it to a certain limit on one side of the 
meridian, when it becomes stationary for a time, and then returns, 
and proceeds to a certain limit on the other side of it, occupying 
two or three centuries in each vibration. At London, in 1580, 
the declination was 11] E. ; in 1657, it was ; after which time 
the needle continued its western movement till 1814, when the 
declination was 24A W. ; since then the needle has been moving 
slowly eastward. The entire secular vibration will probably last 
more than three centuries. The average variation from 1580 to 
1814 was 9 10" annually. But like other vibrations, the motion 
is slowest toward the extremes. 

There has also been detected a small annual variation, in which 
the needle turns its north pole a few minutes to the east of its 
mean position between April and July, and to the west the rest 
of the year. This annual oscillation does not exceed 15 or 18 

372. Diurnal Variation. The needle is also subject to a 
small daily oscillation. In the morning the north end of the 
needle has a variation to the east of its mean position greater than 
at any other part of the day. During winter this extreme point is 
attained at about 8 o clock, but as early as 7 o clock in the sum 
mer. After reaching this limit it gradually moves to the west, 
and attains its extreme position at about 3 o clock in winter, and 
1 o clock in summer. From this time the needle again returns 
eastward, reaching its first position about 10 P. M., and is almost 
stationary during the night. The whole amount of the diurnal 
variation rarely exceeds 12 minutes, and is commonly much less 
than that. These diurnal changes of declination are connected 
with changes of temperature, being much greater in summer than 
in winter. Thus, in England the mean diurnal variation from 
May to October is 10 or 12 minutes, and from November to April, 
only 5 or 6 minutes. 



FIG 218 

373. Dip of the Needle. A needle first balanced on a hor 
izontal axis, and then magnetized and placed in the magnetic meridi 
an, assumes a fixed relation to 

the horizon, one pole or the 
other being usually depressed 
below it. The angle of de 
pression is called the dip of 
the needle. Fig. 218 repre 
sents the dipping needle, with 
its adjusting screws and 
spirit-level; and the depres 
sion may be read on the grad 
uated scale. After the hori 
zontal circle m is leveled by 
the foot-screws, the frame A 
is turned horizontally till the 
vertical circle M is in the 
magnetic meridian. For 
north latitudes, the north 
end of the needle is depressed, 
as a in the figure. 

374. Is o clinic Curves. 

A line passing through all 

points where the dip of the needle is nothing, i. e. where the dip 
ping needle is horizontal, is called the magnetic equator of the 
earth. It can be traced in Fig. 219 as an irregular curve around 

FIG. 219. 

the earth in the region of the equator, nowhere departing from it 
more than about 15. At every place north of the magnetic equa 
tor the north pole of the needle descends, and south of it the south 
pole descends; and, in general, the greater the distance, the 


greater is the dip. Imagine now a system of lines, each passing 
through all the points of equal dip; these will be nearly parallel 
to the magnetic equator, which may be regarded as the standard 
among them. These magnetic parallels are called the isoclinic 
curves ; they somewhat resemble parallels of latitude, but are in 
clined to them, conforming to the oblique position of the magnetic 
equator. In the figure, the broken lines show the dip of the south 
pole of the needle ; the others, that of the north pole. The points 
of greatest dip, or dip of 90, are called the poles of dip. There is 
one in the northern hemisphere, and one in the southern. The 
north pole of dip was found, by Capt. James C. Ross, in 1831, to 
be at or very near the point, 70 14 K ; 96 40 W., marked x in 
the figure. The south pole is not yet so well determined. 

At the poles of dip the horizontal needle loses all its directive 
power, because the earth s magnetism tends to place it in a verti 
cal line, and, therefore, no component of the force can operate in 
a horizontal plane. The isogonic lines in general converge to the 
two dip-poles ; but, for the reason just given, they cannot be traced 
quite to them. 

The dip of the needle, like the declination, undergoes a varia 
tion, though by no means to so great an extent. In the course 
of 250 years, it has diminished about five degrees in London. In 
1820 it was about 70, and diminishes from two to three minutes 

Since the dip at a given place is changing, it cannot be sup 
posed that the poles are fixed points ; they, and with them the en 
tire system of isoclinic curves, must be slowly shifting their lo 

375. Magnetic Intensity of the Earth. The force ex 
erted by the magnetism of the earth varies in different places, 
being generally least in the region of the equator, and greatest in 
the polar regions. The ratio of intensity in different places is 
measured by the number of vibrations which the needle makes in 
a given time. In the discussion of the pendulum, it was proved 
(Art. 170) that gravity varies as the square of the number of vi 
brations. For the same reason, the magnetic force at any place 
varies as the square of the number of vibrations of the needle at 
that place. 

376. Isodynamic Curves. After ascertaining, by actual 
observation, the intensity of the magnetic force in different parts 
of the earth, lines are supposed to be drawn through all those 
points in which the force is the same; these lines are called isody- 
numic curves, represented in Fig. 220. These also slightly re 
semble parallels of latitude, but are more irregular than the 



isoclinic lines. There is no one standard equator of minimum in 
tensity, but there are two very irregular lines surrounding the 
earth in the equatorial region, in some places almost meeting each 

FIG. 220. 

other, and in others spreading apart more than two thousand 
miles, on which the magnetic intensity is the same. These two 
are taken as the standard of comparison, because they are the 
lowest which extend entirely round the globe. The intensity on 
them is therefore called unity, marked 1 in the figure. In the 
wide parts of the belt which they include lying one in the south 
ern Atlantic, and the other in the northern Pacific oceans there 
are lines of lower intensity which return into themselves, without 
encompassing the earth. In approaching the polar regions, both 
north and south, the curves, retaining somewhat the form of 
the unit lines, are indented like an hour-glass, as those marked 
1.7 in the figure, and at length the indentations meet, forming an 
irregular figure 8 ; and at still higher latitudes, are separated into 
two systems, closing up around two poles of maximum intensity. 
Thus there are on the earth four poles of maximum intensity, two 
in the northern hemisphere and two in the southern. The Amer 
ican north pole of intensity is situated on the north shore of Lake 
Superior. The one on the eastern continent is in northern. Si 
beria. The ratio of the least to the greatest intensity on the earth 
is about as 0.7 to 1.9 ; that is, as 1 to 2f. In the figure, intensities 
less than 1 are marked by dotted lines. 

377. Magnetic Charts. These are maps of a country, or of 
the world, on which are laid down the systems of curves which 
have been described. But for the use of the navigator, only the 
isogonic lines, or lines of equal declination, are essential. There 
are large portions of the globe which have as yet been too imper- 


fectly examined for the several systems of curves to be accurately 
mapped. It must be remembered, too, that the earth is slowly 
but constantly undergoing magnetic changes, by which, at any 
given place, the declination, dip, and intensity are all essentially 
altered after the lapse of years. A chart, therefore, which would 
be accurate for the middle of the nineteenth century, will be, to 
some extent, incorrect at its close. 

378. Magnetic Observatories. In accordance with a sug 
gestion of Humboldt, in 183G, systematic observations have been 
since made upon terrestrial magnetism, in various parts of the 
world, in order to deduce from them the laws of its changes. 
Buildings have been erected without any iron in their construc 
tion, to serve as magnetic observatories ; and the most delicate 
magnetometers have been devised and used for detecting minute 
oscillations both in the horizontal and vertical planes. By these 
means has been discovered a class of phenomena called magnetic 
storms, in which the needle suffers numerous and rapid disturb 
ances, sometimes to the extent of several degrees ; and it is a re 
markable and interesting fact that these disturbances occur at the 
same absolute time in every part of the earth. 

379. Aurora Borealis. This phenomenon is usually ac 
companied by a disturbance of the needle, thus affording visible 
indications of a magnetic storm ; but the contrary is by no means 
generally true, that a magnetic storm is accompanied by auroral 
light. The connection of the aurora borealis with magnetism is 
manifested not only by the disturbance of the needle, but also by 
the fact that the streamers are parallel to the dipping needle, as is 
proved by their apparent convergence to that point of the sky to 
which the dipping-needle is directed. This convergence is the 
effect of perspective, the lines being in fact straight and parallel. 

380. Source of the Earth s Magnetism. If a needle is 
carried round the earth from north to south, it takes approxi 
mately all the positions in relation to the earth s axis which it as 
sumes in relation to a magnetic bar, when carried round it from 
end to end (Art. 367). At the equator it is nearly parallel to the 
axis, and it inclines at larger and larger angles as the distance 
from the equator increases ; and in the region of the poles, it is 
nearly in the direction of the axis. The earth itself, therefore, 
may be considered a magnet, since it affects a needle as a magnet 
does, and also induces the magnetic state on iron. But it is nec 
essary, on account of the attraction of opposite poles, to consider 
the northern part of the earth as being like the south pole of a 
needle, and the southern part like the north pole. To avoid this, 


the words boreal and austral are applied to the two magnetic 
states, and the boreal magnetism is the name given to that devel 
opment found in the northern hemisphere, and the austral mag 
netism to that in the southern. Hence, it becomes necessary, in 
using these names for a magnet, to reverse their order, and to 
speak of its north pole as exhibiting the austral, and its south pole 
the boreal magnetism. 

Modern discoveries in electro-magnetism and thermo-electri 
city furnish a clew to the hypothesis which generally prevails at 
this day. Attention has been drawn to the remarkable agreement 
between the isothermal and the isomagnetic lines of the globe. 
The former descend in crossing the Atlantic Ocean toward Amer 
ica, and there are two poles of maximum cold in the northern 
hemisphere. The isoclinic and the isodynamic curves also de 
scend to lower latitudes in crossing the Atlantic westward ; so 
that, at a given latitude, the degree of cold, the magnetic dip, and 
the magnetic intensify, is each considerably greater on the Amer 
ican than on the European coast. This is only an instance of the 
general correspondence between these different systems of curves. 
It has likewise been noticed (Art. 372) that the needle has a move 
ment diurnally, varying westward during the middle of the day, 
a.nd eastward at evening, and that this oscillation is generally 
much greater in the hot season than the cold. It is obvious, 
therefore, that the development of magnetism in the earth is inti 
mately connected with the temperature of its surface. Hence it is 
supposed that the heat received from the sun excites electric cur 
rents in the materials of the earth s surface, and these give rise to 
the magnetic phenomena. 

381. Formation of Permanent Magnets. Needles and 
small bars may be more or less magnetized by the following meth 
ods, the reasons for which will be readily understood : 

1. A feeble magnetism may be developed in a steel bar, by 
causing it to ring while held vertically. The earth s influence 
upon it, however, is stronger if it is held, not precisely vertical, 
but leaning in a direction parallel to the dipping needle. The 
inductive influence of the earth explains the fact often noticed, 
that rods of iron or steel that have stood for many years in a posi 
tion nearly vertical, as, for instance, lightning-rods, iron pillars, 
stoves, &c., are found somewhat magnetic, with the north pole 

2. A needle may be magnetized by simply suffering it to re 
main in contact with the pole of a strong magnet, or better, be 
tween the opposite poles of two magnets. 

3. Place the needle across the opposite poles of two parallel 



magnets, while a bar of soft iron connects the other two poles. 
Thus, removing one of the keepers, A, B, from the ends of the 
magnets (Fig. 221), put the needle in its place, being careful that 
the end of the needle 
marked for north is ad 
jacent to the south pole 
of the magnet. 

4. In order to take 
advantage of the earth s 
inductive influence, along with that of steel magnets, place the 
needle parallel to the dipping needle, and draw the south pole of 
one magnet over the lower half, and the north pole of another 
over the upper half, with repeated and simultaneous move 

None of these methods, however, are of great practical value at 
the present day, since the galvanic circuit affords a far readier and 
more efficient means of magnetizing bars. 

The horse-shoe magnet, sometimes called the FIG. 222. 
U-magnet (Fig. 222), is for many purposes a very 
convenient form, and originated in the practice of 
arming the lodestone; that is, furnishing it with two 
pieces of soft iron, which are confined by brass straps 
to the poles of the stone, and project below it, so 
that a bar and weight may be attached. When a 
magnet has this form, both poles may be applied to a 
body at once. The C7-magnet, A N 8, being sus 
pended, and the keeper, B, made of soft iron, being 
attached to the poles, weights may be hung upon 
the hook C, to show the strength. 

382. The Declination Compass. This in 
strument consists of a magnetic needle suspended in 
the centre of a cylindrical brass box covered with 
glass ; on the bottom of the box within is fastened a 
circular card, divided into degrees and minutes, from 
to 90 on the several quadrants. On the top of ^ 

the box are two uprights, either for holding sight- 
lines or for supporting a small telescope, by which directions are 
fixed. The quadrants on the card in the box are graduated from 
that diameter which is vertically beneath the line of sight. 

When the axis of vision is directed along a given line, the 
needle shows how many degrees that line is inclined to the mag 
netic meridian. In oVder that the angle between the line and the 
geographical meridian maybe found, the declination of the needl- 
for the place must be known. 


383. The Mariner s Compass. In the mariner s compass 
(Fig. 223) the card is made as light as possible, and attached to the 
needle, so that the north and south 
points marked on the card always 
coincide with the magnetic merid 
ian. The index, by which the di 
rection of the ship is read, consists 
of a pair of vertical lines, diametri 
cally opposite to each other, on the 
interior of the box. These lines, 
one of which is seen at #, are in the 
plane of the ship s keel. Hence, 
the degree of the card which is against either of the lines shows 
at once both the angle with the magnetic meridian and the quad 
rant in which that angle lies. 

In order that the top of the box may always be in a horizontal 
position, and the needle as free as possible from agitation by the 
rolling of the ship, the box, B, is suspended in gimbals. The 
pivots, A, A, on opposite sides of the box, are centred in the brass 
ring, C f D, while this ring rests on an axis, which has its bearings 
in the supports, E, E. These two axes are at right angles to each 
other, and intersect at the point where the needle rests on its 
pivot. Therefore, whatever position the supports, E, E, may have, 
the box, having its principal weight in the lower part, maintains 
its upright position, and the centre of the needle is not moved by 
the revolutions on the two axes. 

On account of the dip, which increases with the distance from 
the equator, and is reversed by going from one hemisphere to the 
other, the needle needs to be loaded by a small adjustable weight, 
if it is to be used in extensive voyages to the north or south. In 
north latitudes the south end must be heaviest; in. south lati 
tudes, the north end. 

384. The Needle Rendered Astatic. Though magnetic 
intensity increases at greater distances from the equator, yet the 
directive power of the compass grows more feeble in approaching 
the poles of dip, because the horizontal component constantly di 
minishes, and at the poles becomes zero (Art. 374). A needle in 
such a situation, in which the earth s magnetism has no influence 
to give it direction, is called astatic. The compass needle is astatic 
at the north and south poles of dip. And the dipping needle 
may be rendered astatic at any place by setting its plane of rota 
tion perpendicular to its line of dip at that*place ; for then there 
will remain no component of the magnetic force- in the only plane 
in which the needle is at liberty to move. 


The needle may also be made astatic at any place by holding a 
magnet at such a distance, and in such a 
position, as to neutralize the earth s in- 
iluence. Or, if a wire, suspended verti 
cally by a thread, pass through the centres 
of two needles, whose poles point in oppo 
site directions, each needle will be astatic. 
The needles in Fig. 224, with like poles in 
opposite directions, are slipped tightly upon 
the wire b c, which is suspended by the 
thread a b, free from torsion. This method 
of liberating a magnetic needle from the 
earth s influence is of great use in electro-magnetism. 

385. Theory of Magnetism. The nature of the agency 
called magnetism is unknown. Much of the language employed 
by writers on the subject implies that there exist in iron, steel, 
&c., two imponderable fluids, called the austral and boreal magnet 
isms ; that these fluids attract each other, and are ordinarily 
mingled and neutralized, so that no magnetic phenomena appear; 
and that in every magnet the two fluids have been separated by 
the inductive influence of the earth or of another magnet, one 
fluid manifesting itself at one pole, and the other at the other 
pole. As science advances, however, these views seem more and 
more crude and unsatisfactory. Magnetism is now regarded by 
many as one of those modes of molecular motion which are so diffi 
cult of investigation. If it is a mode of motion, then it may man 
ifest itself as a force, as we know it does. It will be seen in the 
discussion of Electro-magnetism that there is a most intimate 
connection between magnetism and electricity, so much so that 
the former is generally considered as only a particular form in 
which the latter is developed. 

Magnetism differs from the other molecular agencies elec 
tricity, light, and heat in producing no direct effect on any of 
our senses. We witness its direct effects only in the motion which 
it gives to certain kinds of matter, such as iron and steel. 





386. Definitions. The name Electricity, from the Greek 
word for amber, is given to a peculiar agency, which is the cause 
of a variety of phenomena, such as attracting and repelling light 
bodies, producing light, heat, sound, and chemical decomposition, 
and, when concentrated in its action, violently rending or explod 
ing bodies. Lightning and thunder are an example of its intense 

Frictional electricity is so called because generally excited by 
friction, and to distinguish this form of development from the gal 
vanic electricity which is excited by chemical means. The former 
is often called statical, and the latter dynamical electricity. 

Bodies are said to be electrically excited when they show signs 
of electricity by some action performed upon them, as friction, for 
example. They are said to be electrified when they receive elec 
tricity by communication. 

Conductors are bodies which transmit electricity freely ; non 
conductors are those which do not transmit it at all, or only very 
imperfectly. A body is said to be insulated when in contact only 
with non-conductors, so that electricity is retained in it. 

387. Electroscopes. The feeblest indication of electricity 
is usually attraction or repulsion ; and instruments prepared for 
showing these effects are called electroscopes. The w T ord electrome 
ter, though sometimes used in the same sense, is more properly 
defined to be an instrument for measuring the quantity of elec 

The pendulum electroscope (Fig. 225) consists of a glass stand 
ard, supported by a base, and bent into a hook at the top, from 
which is suspended a pith ball by a fine silk thread. 


The gold-leaf electroscope consists of two narrow strips of gold- 
leaf, n, n (Fig. 22G), suspended within a glass receiver, B, from a 

FIG. 235. 


metallic rod which passes through the top and terminates in a 
ball, (7. A metallic base is cemented to the receiver, and strips of 
tin-foil, a, are attached to the inside, reaching to the base. When 
an electrified body is brought near the knob C 9 the gold leaves 
separate, or, if separated, collapse, or separate more, according to 

Certain modifications are convenient for some purposes. One 
is, a metallic wire with a ball on the top, having a thread and 
pith ball hanging by the side of it; and another, two threads with 
pith balls suspended together below a conductor, us in Fig. 231. 

388. Common Indications of Electricity. Though the 
frictional or statical electricity may be developed in several ways, 
pressure, evaporation, &c., the method generally employed is fric 
tion. By this means it can be excited in a greater or less degree 
in all substances, and from some it may be easily and abundantly 

If amber, sealing-wax, or any other resinous substance, be 
rubbed with dry woolen cloth, fur, or silk, and then brought near 
the face, the excited electricity disturbs the downy hairs upon the 
skin, and thus causes a sensation like that produced by a cobweb. 
When the tube is strongly excited, it gives oif a spark to the finger 
held toward it, accompanied by a sharp snapping noise. A sheet 
of writing-paper, first dried by the fire, and then laid on a table 
and rubbed with india-rubber, becomes so much excited as to ad 
here to the wall of the room or any other surface to which it is 
applied. As the paper is pulled up slowly from the table by one 
edge, a number of small sparks may be seen and heard on the 


under side of the paper. In dry weather, the brushing of a gar 
ment causes the floating dust to fly back and cling to it. 

If an iron or brass rod be held in the hand and rubbed with 
silk, the rod shows no sign of electricity. It will be seen here 
after that the electricity excited in the rod is conveyed away by 
the conducting quality of the metal and the human body. 

389. The Two Electrical States. When friction has 
taken place between two bodies, they are found in electrical condi 
tions, which in some remarkable particulars are unlike each other. 
These two electrical states are usually called the positive and the 
negative, terms which were employed by Franklin in his theory of 
one electric fluid, to indicate that the excited body has either more 
or less electricity than belongs to it in its common unexcited con 
dition. Du Fay, in his theory of two kinds of electricity, uses the 
words vitreous and resinous to distinguish them, vitreous corre 
sponding to the positive, and resinous to the negative. But it is 
very common to use Du Fay s theory, and to apply Franklin s 
terms, positive and negative, to the two kinds of electricity. 

If in any case only one electricity is discovered when friction 
causes development, it is to be understood that the other is dif 
fused through some large conductor, so as to be imperceptible. 
The earth is the great reservoir, in which any amount of elec 
tricity may be diffused and lost sight of. 

390. Nature of Electricity. The real nature of electricity 
is unknown. Though it is in most treatises spoken of as a fluid, 
of exceeding rarity, and more rapid in its movements than light, 
yet the prevailing belief at the present day is, that it is a peculiar 
mode of vibratory motion, either in the luminiferous ether which 
is imagined to fill all space, or else in the ordinary matter consti 
tuting the bodies and media about us, or in both of these. Elec 
tricity is brought to view by friction, by heat, and by other 
agencies which are calculated to cause movements in matter, rather 
than to bring new kinds of matter to light. It is undoubtedly one 
of the forms of force, into which other forces may be transformed. 
But until a more definite ivave-theory or force-theory can be con 
structed than exists at present, it is comparatively easy to give to 
the learner an intelligible description of electrical phenomena by 
using the language of the two-fluid theory of Du Fay. In trying 
to give a statement of observed facts without the use of these hy 
pothetical terms, it is necessary to employ in their stead tedious 
circumlocutions, which only confuse the mind of the learner. 

391. Du Fay s Theory. According to this theory, the two 
fluids are imagined to inhere in all kinds of matter, combined with 


each other and neutralized. In this condition, they afford no evi 
dence of their existence. But they can in several ways be scjxt- 
rated from each other; and when thus separated, they give rise to 
electrical phenomena. 

392. The Two States Developed Simultaneously. 

If bodies are rubbed together, the two electricities are separated, 
and one body is electrified positively, the other negatively. For 
example, glass rubbed with silk is itself positive, and the silk is 
negative. But the same substance does not always show the same 
kind of electricity, since that depends frequently on the substance 
against which it is rubbed. Dry woolen cloth rubbed on smooth 
glass is negative, but on sulphur it is positive. The following 
table contains a few substances, arranged with reference to this. 
Any one of them, rubbed with one that follows it, is positively 
electrified itself, and the other negatively : 

1. Fur of a cat. 7. Silk. 

2. Smooth glass. 8. Gum lac. 

3. Flannel. 9. Resin. 

4. Feathers, 10. Sulphur. 

5. Wood. 11. India-rubber. 

6. Paper. 12. Gutta-percha. 

According to the above table, silk rubbed on smooth glass is 
negatively excited ; but rubbed on sulphur, it is excited positively. 
It is sometimes found, however, that the previous electrical condi 
tion of one of the bodies will invert the order stated in the table. 
For example, if silk, having been rubbed on smooth glass, and 
therefore being negative, should then be rubbed on resin, it would 
probably retain its negative state, and the resin become positively 
electrified, contrary to the order of the table. 

The mechanical condition of the surface sometimes changes 
the order of the two electricities. Thus, if glass is ground, so as 
to lose its polish, it is likely to be negative when rubbed with 
silk ; but the excitation of rough glass is very feeble. 

393. Mutual Action. Bodies electrified in different ways 
attract, and in the same way repel each other. Thus, if an insu 
lated pith ball, or a lock of cotton, be electrified by touching it 
with an excited glass tube, it will immediately recede from the 
tube, and from all other bodies which are charged with the posi 
tive electricity, while it will be attracted by excited sealing-wax, 
and by all other bodies which are negatively electrified. If a lock 
of fine long hair be held at one end, and brushed with a dry brush, 
the separate hairs will become electrified, and will repel each other. 
In like manner, two insulated pith balls, or any other light bodies, 


will repel each other when they are electrified the same way, and 
attract each other when they are electrified in different ways. 

Hence it is easy to determine whether the electricity developed 
in a given body is positive or negative ; for, having charged the 
electroscope with excited glass, then all those bodies which, when 
excited, attract the ball, are negative, while all those which repel 
it are positive. 

394. Conduction. Electricity passes through some bodies 
with the greatest facility; through others with difficulty, or 
scarcely at all; and others still have a conducting power interme 
diate between the two. As the conducting quality exists in differ 
ent substances in all conceivable degrees, it is impossible to draw 
a dividing line between them, so as to arrange all conductors on 
one side, and all non-conductors on the other. The following 
brief table contains some of the more important of the two classes ; 
the first column in the order of conducting power, the second in 
the order of insulating power: 

Conductors. Insulators. 

The metals, Lac, amber, the resins, 

Charcoal, Paraffine, 

Plumbago, Sulphur, 

Water, damp snow, Wax, 

Living vegetables, Glass, precious stones, 

Living animals, Silk, wool, hair, feathers, 

Smoke, steam, Paper, 

Moist earth, stones, Air, the gases, 

Linen, cotton. Baked wood. 

When air is rarefied, its insulating power is diminished, and 
the further the rarefaction proceeds, the more freely does elec 
tricity pass. Hence, we might expect that it would pass with per 
fect freedom through a complete vacuum. It is found, however, 
that in an absolute vacuum electricity cannot be transmitted 
at all. 

395. Modes of Insulating. Solid insulating supports are 
usually made of glass ; and, in order to improve their insulating 
power, they are sometimes covered with shell-lac varnish. Insu 
lating threads for pith balls, or cords for suspending heavier 
bodies, are made of silk. The best insulator for suspending any 
yery small weight is a single fiber of silk, a hair, or a fine thread 
of gum lac. In order to perform electrical experiments, the air 
must be dry, or no care whatever relating to apparatus can insure 
success ; and therefore, in a room occupied by an audience, es 
pecially if the weather is damp, it is necessary to dry the air arti- 


ficially by fires. If the air were a good conductor, it is probable 
that 110 facts in this science would ever have been discovered. 

396. Communication and Influence. The sp?iere of com- 
munication is the space within which a spark may pass from an 
electrified body, in any direction. It is sometimes called the 
striking distance. The sphere of influence is the space within 
which the power of attraction of an electrified body extends every 
way, beyond the sphere of communication. A glass tube strongly 
excited will give motion to the gold-leaf electroscope at the dis 
tance of several feet, although a spark could not pass from the 
tube to the cap of the electroscope at a greater distance than a few 
inches. The electricity which a body manifests by being brought 
towards an excited body, without receiving a spark from it, is said 
to be acquired by induction. The principle of induction resembles 
that noticed in magnetism, and will be discussed in connection 
with the Leyden jar. 



397. The Plate Machine. In order that glass may be con 
veniently subjected to friction for the development of electricity, 
it is made in the form of a circular plate, and mounted on an axis, 
which is supported by a wooden frame, and revolved by a crank, 
while rubbers press against its surface. Fig. 227 represents one 
of the many forms which have been adopted. The crank, M, gives 
rotary motion to the plate, P, which is pressed by the rubbers, 
F, Fy this pressure is equalized by their being placed at top and 
bottom, and on both sides of the glass. The prime conductor, 
C C, is made of hollow brass, and supported by glass pillars. The 
extremities terminate in two bows, which pass around the edges 
of the plate, and present to it a few sharp points, to facilitate the 
passage of electricity. But all other parts are carefully rounded 
in cylindrical and spherical forms, without edges or points, as 
these tend to dissipate the electricity. The glass, as it revolves 
from the rubbers to the points of the prime conductor, is pro 
tected by silk covers, to prevent the electricity from escaping into 
the air. The rubbers are made of soft leather, attached to a piece 
of wood or metal, and from time to time are rubbed over with an 
amalgam of zinc, tin, and mercury, or with the bi-sulphuret of 



tin, which is one of the best exciters on glass. The diameter of 
the plate varies from 1^ to 3 feet; but in some of the largest it is 
6 feet, and two plates are sometimes mounted on one axis. 

FIG. 227. 

To give free passage of the negative electricity from the rub 
bers to the earth, a chain, D, may be attached to the wooden sup 
port, while its other end lies on the floor. 

398. The Cylinder Machine. In many electrical machines 
of the smaller sizes, a hollow cylinder is employed, having a length 
considerably exceeding its diameter. In the cylinder machine, 
the rubber is applied to one side, and the prime conductor receives 
the fluid from the opposite. The rubber is usually mounted on a 
glass pillar, so that it can be insulated, whenever it is desired 

399. The Hydro-Electric Machine. It was discovered in 
1840 that a steam-boiler electrically insulated gave out sparks, 
and that the steam issuing from it was also electrified. Hence re 
sulted the construction of the hydro-electric machine. It consists 
of a boiler mounted on glass pillars, and furnished with a row of 


jet-pipes and a metallic plate, against which the steam strikes. 
The prime conductor, to which the steam-plate is attached, is 
electrified positively, and the boiler itself negatively. Professor 
Faraday ascertained that the electricity in this case is developed, 
not by evaporation or condensation, but by the friction of watery 
particles in the jet-pipes. That the machine may act with energy, 
it was found necessary to make the interior of the jet-pipes angu 
lar, and quite irregular. 

In connection with the subject of induction will be described 
a machine of still more recent invention, and known as the induc 
tion machine. 

400. The Quadrant Electrometer. In order to measure 
the intensity of electricity in the prime con 
ductor, there is set upon it, whenever desired, 

a quadrant electrometer (Fig. 228). This con 
sists of a pillar, d, about six inches high, having 
a graduated semicircle, c, attached to one side, 
and a delicate rod and ball, a, suspended from 
the centre of the semicircle. As the conductor 
becomes electrified, the rod is repelled from the 
pillar, and the arc passed over indicates rudely 
the degree of electrical intensity. 

401. First Phenomena cf the Ma 
chine. When an electrical machine is skill 
fully fitted up, and works well, there is first 

perceived, on turning it, a crackling sound; and then, on bring 
ing the knuckles toward the prime conductor, a brilliant spark 
leaps across, causing a sharp pricking sensation. If the room be 
darkened, brushes of pale light are seen to dart off continually 
from the most slender parts of the prime conductor, with a hiss 
ing or fluttering noise, while circles of light snap along the glass 
between the rubbers and. the edges of the covers. When electricity 
is escaping plentifully from the machine, a person standing near 
also perceives a peculiar odor, which is that of ozone, and which 
seems always to accompany the development of electricity. 

Therefore, at least four of the senses are directly affected by 
this remarkable agency, while magnetism affects none of them. 

The phenomena of repulsion of like and attraction of unlike 
electricities, are well shown by the machine. A skein of thread 
or a tuft of hair, suspended from the prime conductor, will, as 
soon as^the plate is revolved, spread into as wide a space as possi 
ble, by the repellency of the fibers which are electrified alike. 
Melted sealing-wax is thrown off in fine threads, and dropping 



v* r ater is diverged into delicate filaments. Even air, on those parts 
of the prime conductor which are most strongly charged, becomes 
so self-repellent as to fly oif in a stream of wind, which is plainly 

On the other hand, light bodies, when brought toward the ma 
chine while in action, instantly fly to the prime conductor ; for 
that is positive, but the nearer sides of the other bodies are made, 
negative by induction. 

The difference between substances as to their conducting qual 
ity is readily perceived by setting the quadrant electrometer on 
the prime conductor, raising the index by turning the plate, and 
then touching the prime conductor with the remote end of the 
body to be tried. If an iron rod, or even a fine iron wire, be thus 
applied, the index will fall instantly; a long dry wooden rod 
will cause it to descend slowly, while a glass rod will produce no 
effect at all. These experiments show that iron is a perfect con 
ductor, wood an imperfect conductor, and glass a non-conductor. 

402. Coulomb s Torsion Balance. When a long fine wire 
is stretched by a small weight, its elasticity of torsion is a very 
delicate force, which is successfully employed for the measurement 
of other small forces. When such a wire is twisted through differ 
ent angles, the force of torsion is found to vary as the angle of tor 
sion ; it is therefore easy to measure the 

force which is in equilibrium with tor 
sion. The torsion balance is represented 
in Fig. 229. The needle of lac, n o, is 
suspended by a very fine wire from a 
stem at the top of the tube d. The cap 
of the tube, e, is a graduated circle, 
whose exact position is marked by the 
index, a. The stem from which the 
wire hangs is held in place in the centre 
of the cap by friction, but can be turned 
round so as to place the needle in any 
direction desired. At the end of the lac 
needle is a small disk of brass-leaf, n, 
and by its side a gilt ball, m, connected 
with the handle, r, by the glass rod, i. 
This apparatus is suspended in the glass 
cylinder, covered with a glass plate, on 

the centre of which the tube d is fastened. There is a graduated 
circle around the c}<linder on the level of the needle. 

403. Law of Electrical Force as to Distance. Adjust 
ment is now made by turning the stem so that, while the wire is 

FIG. 229. 


in its natural condition, the disk, n, touches the ball, m, and is at 
zero, and the index at top also at zero on the circle e. Let a mi 
nute charge of electricity be communicated to m, and it will repel 
#, and cause it, after a few oscillations, to settle at a certain dis 
tance suppose, for instance, at 36. The circle e is now turned in 
the opposite direction, until the needle is brought within 18 of 
the ball m. In order to bring it thus near, the index has to be 
turned 126, which added to the 18, makes the whole torsion 
144, or four times as great as before. Therefore, at one-half the 
distance there is four times the repulsion. In like manner, it is 
found that at one-third the distance there is nine times the repul 
sion. Hence, the law, 

Electrical repulsion varies inversely as the square of the dis 

In a manner somewhat similar to the foregoing, it was conclu 
sively proved by Coulomb that electrical attraction obeys the same 
law of distance, though there is more practical difficulty in per 
forming the experiments. But if the electrified body m is placed 
outside of the circle described by n, so that the latter is allowed to 
vibrate both to the right and left, the square of the number of 
vibrations in a given time becomes a measure of the attractive 
force, as in the case of the pendulum (Art. 170). 

404. Waste of Electricity from an Insulated Body. 

In making accurate investigations like the foregoing, in which 
considerable time is necessarily occupied, a difficulty arises from 
the loss of the electrical charge. The first and most obvious 
source of waste is the moisture in the air, which conducts away 
the fluid ; but this may be nearly avoided by setting into the cyl 
inder a cup of dry lime, or other powerful absorbent of moisture, 
as represented in the figure. A second is the imperfect insulation 
afforded by even the most perfect non-conductors. A third is the 
mobility of the air, whose particles, when they have touched the 
electrified body, and become charged, are repelled, taking away 
with them the charge they have received. The loss in these ways 
is very slight, when the charge is small, and allowance can be 
made for it with a good degree of accuracy. But when bodies are 
highly charged, they lose their electricity at a rapid rate. 

405. An Electrical Charge Lies at the Surface. This 
is proved in many ways. A hollow ball, no mutter how thin, will 
receive as large a quantity of electricity as a solid one. Hence it 
is that the prime conductor of the electrical machine, and metallic 
articles of electrical apparatus generally, are made of sheet brass, 
for the sake of lightness. 

Let a metallic ball, supported on a glass pillar, be charged 


with either kind of electricity. Then apply to it two thin metal 
lic hemispheres, by means of insulating handles. If they now be 
quickly removed from the ball, all the electricity which was pre 
viously on the ball is found On the hemispheres. 

Let a dish, a (Fig. 230), be made of two brass rings and cam 
bric sides and bottom, with an insulating 
handle, &, attached to the larger ring. If FlG - 

this vessel be charged with electricity, the 
charge is found on the outside ; turn it 
over quickly, so as to throw it the other 
side out, and the charge is instantly found 
on the outside again, and none on the inside. It may be inverted 
several times with the same result, before the charge becomes too 
feeble to be perceived. 

If cavities are sunk into a solid conductor, no sensible quantity 
of electricity is found at the bottom of such cavities. In experi 
ments of this kind, Coulomb found his torsion balance (Fig. 229) 
of great service. A proof plane, as he termed it that is, a small 
piece of gilt paper cemented upon the end of a slender rod of lac, 
was first touched to that part of an electrified body which was to 
be examined, and then applied to the ball of the instrument. The 
distance to which the needle was repelled indicated the intensity 
of electricity at the point in question. The charge taken from the 
bottom of an abrupt cavity was never sufficient to move the 

Another proof that the charge occupies only the outside sur 
face is that the intensity diminishes as the surface is enlarged, 
while the mass of the conductor remains the same. A metallic 
ribbon rolled upon an insulated cylinder may be unrolled, and 
thus the surface enlarged to any extent. An electroscope standing 
on the instrument will fall as the ribbon is unrolled, and rise 
when it is again rolled up. 

406. Distribution of a Charge on the Surface. Devel 
oped electricity resides at the surface of a body, as we have seen, 
but is not uniformly diffused over it, except in the case of the 
sphere. In general, the more prominent the part, and the more 
rapid its curvature, the more intensely is the fluid accumulated 

In a long slender rod, nearly the whole charge is collected at 
the extremities. On the surface of an ellipsoid it is found to be 
arranged according to a very simple law, namely: the quantity of 
the charge at each point varies as the diameter through that point. 
But the tendency to escape increases at a more rapid rate, and 
varies as the square of the diameter. Hence it is that electricity 


is so rapidly dissipated from points, which may be regarded as the 
extremities of ellipsoids indefinitely elongated. If the surface of a 
bodv is partly convex and partly concave, the distribution is still 
more unequal ; nearly all the charge collects on the convex parts ; 
and if the concavities are deep or abrupt, like those mentioned in 
Art. 405, no sign of electricity is discovered in them. 

407. Rotation by Unbalanced Pressure. As the electric 
charge on the surface of a body presses outward in all directions, 
wherever it escapes from a point, there the pressure is removed ; 
consequently, on the opposite part there is unbalanced pressure. 
Therefore, if the body is delicately suspended, and one or more 
points are directed tangentially, the unbalanced pressure will 
cause rotation in the opposite direction, just as Barker s mill ro 
tates by the unbalanced pressure of water. Electrical wheels and 
orreries are revolved in this way. 

A windmill may also be revolved by the stream of air issuing 
from a stationary point attached to the prime conductor (Art. 401). 

408. The Charge Hsfd on the Surface by Atmospheric 
Pressure. The mutual repellency, which drives the particles 
asunder till they reach the surface of the conductor, tends to make 
them escape in all directions from that surface ; and it is the air 
alone which prevents. For if one extremity of a charged and in 
sulated conductor extends into the receiver of an air-pump, the 
charge is dissipated by degrees, as the receiver is exhausted ; and 
when the exhaustion is as complete as possible, the mgst abundant 
supply from the machine fails to charge the conductor. As the 
atmospheric pressure is limited to about 15 Ibs. per square inch, so 
the amount of charge is limited which can be retained on a con 
ductor of given form. Hence the reason for the well-known fact 
that the prime conductor receives all the charge which it is capa 
ble of retaining in one or two turns of the machine. All that is 
gained over and above this, by continuing to turn, flies off through 
the air. 



409. Elementary Experiment, When an electrified body 
is placed near one which is unelectrified, but not within the 
sphere of communication, the natural electricities of the latter are 


decomposed, one being attracted toward the former, the other re 
pelled from it (Art. 396). Thus the ends become electrified by 
the influence of the first body, without receiving any electricity 
from it. Let A (Fig. 231) be charged with positive electricity, 

FIG. 231. 

A A 1 A A 

and let the insulated conductor, B C, be furnished with several 
electroscopes, as represented. Those nearest the ends will diverge 
most, and the others less according as they are nearer the centre, 
where there is no sign of electricity. By taking off small quanti 
ties with the proof-plane, and testing them, it is found that nega 
tive electricity occupies the end nearest to A 9 and positive the 
remote end. Eemove the bodies to a distance from each other, 
and B C returns to its unelectrified condition ; bring them near 
again, and it is electrified as before. As this electrical state is in 
duced upon the conductor by the electrified body in its vicinity, 
without any communication of electricity, it is said to be electri 
fied by induction. If A is first charged with the negative elec 
tricity, the two electricities of B C will be arranged in reversed 
order ; the positive will be attracted to the nearest end, the nega 
tive repelled to the farthest. 

Electrical induction is exactly analogous to magnetic induc 
tion ; the opposite kind is developed at the nearer end, and the 
like kind at the remote end. 

410. Successive Actions and Reactions. If A is itself 
an insulated conductor, the foregoing is not the entire effect ; for 
a reflex influence is exerted by the electricity in the nearer end of 
the conductor. Let A have a positive charge, as at first. After 
the negative electricity is attracted to the nearer end of B C, it in 
turn attracts the positive charge of A, and accumulates it on the 
nearest side, leaving the remote side less strongly charged than 
before. This is shown by electroscopes attached to the opposite 
sides of A. The charge of A, being now nearer, will exert more 
power on B C, separating more of its original electricities, and 


thus making the nearest end more strongly negative and the re 
mote end more strongly positive than before ; and this new ar 
rangement of fluids in B C causes a second reaction upon A, of 
the same kind as the first Thus ail indefinite diminishing series 
of adjustments takes place in a single moment of time. 

411. Division of the Conductor. Suppose that hcfore the 
experiment begins, B C is in two parts with ends in contact; the 
entire series of mutual actions takes place as already described. 
Now, while A remains in the vicinity, let the parts of B C be sep 
arated ; then the negative electricity is secured in the nearest half, 
and the positive in the other. And if A is now removed, the pos 
itive charge is diffused over the more distant half. Thus each 
kind of electricity can be completely separated from the other by 
means of induction. 

Here we find a marked difference between magnetism and fric- 
tional electricity. The electricities may be secured in their sep 
arate state, one in one conductor, the other in another. In mag 
netism this is not possible ; for when an iron bar is magnetized, 
and then broken, each kind of magnetism is found in each half of 
the bar. At the point of division both polarities exist, and as soon 
as the bar is broken, they manifest themselves there as strongly as 
at the extremities. 

412. Effect of Lengthening the Conductor. If the con 
ductor, B C, is lengthened, the accumulation on the adjacent parts 
of the two bodies is somewhat increased. The positive electricity 
which, at the remote end of the shorter conductor, operated in some 
degree .by its repulsion to prevent accumulation on the nearest 
side of A, is now driven to a greater distance ; and therefore a 
larger charge will come from the remote to the nearer side of A, 
which in turn attracts more negative to the nearer end of B C, 
and thus a new series of actions and reactions takes place in addi 
tion to the former. To obtain the greatest effect from this cause, 
the conductor, B (7, is connected with the earth that is, it is un 
insulated; then the positive part of its decomposed electricities is 
driven to the earth, and entirely disappears, and the negative part 
is attracted to the nearer end ; so that, when the series of adjust 
ments is completed, the remote end of the conductor is in the 
neutral state. This experiment is performed by touching the 
finger to the conductor, after it has become electrified by induc 
tion. The electroscope nearest to A instantly rises a little higher, 
and the distant ones collapse. 

If the original charge in A was negative instead of positive, the 
foregoing experiments are in all particulars the same, except that 
the order of the two fluids is reversed. 


413. Disguised Electricity. The electricity which occu 
pies the surface of the prime conductor, or any other body electri 
fied in the ordinary way, and which is kept from diffusing itself 
in every direction only by the pressure of the air (Art. 408), is 
called free electricity ; for it will instantly spread over the surface 
of other conductors, when they are presented, and therefore will 
be lost in the earth, the moment a communication is made. But 
the electricity which is accumulated by the inductive influence is 
not free to diffuse itself; the same attractive force which has con 
densed it still holds it as near as possible to the original charge ; 
and if we touch the electrified body with the hand, the electricity 
does not pass off; it is therefore called disguised electricity. In 
this respect the two fluids on the contiguous sides of A and B 
are alike ; either may be touched, or in any way connected with 
the earth, but, unless communication is made between them, or 
unless they are both allowed to pass to the earth, they hold each 
other in place by their mutual attraction, and show none of the 
phenomena of free electricity. 

414. A Series of Conductors. If another insulated con 
ductor, D, is placed near to the remote end of B C, and A is 
charged positively, then that extremity of B C nearest to D is in 
ductively charged with positive, as already stated. Hence, the 
electricities of D are separated, the negative approaching B C, and 
the positive withdrawing from it ; there is therefore the same ar 
rangement of fluids in both bodies, but a less intensity in D than 
in B C. For, on account of distance, the positive is not so in 
tensely accumulated at the remote end of B C as in the original 
body A, and therefore a less force operates on D than on B C. 
The same effects are produced in a less and less degree in an in 
definite series of bodies ; and the shorter they are, the more nearly 
equal will be the successive accumulations. The same facts were 
noticed in a series of magnets. 

415. An Electrified Body Attracts an Unelectrified 
Body. This fact, which is the first to be noticed in observing 
electrical phenomena (Art. 401), is explained by induction. If 
B C is light, and delicately suspended, a consequence of the ar 
rangement of fluids already described is, that B C will move to 
ward A. For, according to the law of distance (Art. 403), the 
negative in the nearer part is attracted more strongly than the 
positive in the remote part is repelled ; hence the body yields to 
the greater force, and moves toward A. That the attracted fluid 
does not leave the body, B C, behind, and go to A, is owing to the 
fact, noticed in Art. 408, that the body and the electricity are con 
fined to each other by atmospheric pressure. 



FIG. 232. 

416. The Inductive Action Greatly Increased. In llv. 

experiments as now described, the inductive influence is feeble, 
and the accumulation of electricities very small ; for the bodies 
present toward each other only a limited extent of area, and they 
are necessarily :is much as four or live inches distant, in order to 
prevent the fluid from passing across. By giving the bodies such 
a form that a largo extent of surface may be equidistant, and then 
interposing a solid non-conductor, as glass, between them, so that 
the distance may be reduced to one-eighth of an inch or less, it is 
easy to increase the attracting and repelling forces many thousands 
of times. Let a glass plate, C D (Fig. 232), supported on a base, 
have attached to the middle of each 
side a rectangular piece of tin-foil. 
This is called a Franklin plate. Let 
A be connected with one coating, and 
B with the other. If, now, A forms 
a part of the prime conductor of an 
electrical machine, and B has com 
munication with the earth, we are 
prepared to notice the remarkable 
phenomena of the Leyden jar. If 
the amount of surface and the thick 
ness of glass are the same, the partic 
ular form of the instrument is im 
material; but, for most purposes, a 
vessel or jar is more convenient than a pane of glass of equal sur 
face, and is generally employed for electrical experiments. 

417. The Leyden Jar. This article of electrical apparatus 
consists of a glass jar (Fig. 233), coated on both sides with tin-foil, 
except a breadth of two or three inches near the top, 

which is sometimes varnished for more perfect insula 
tion. Through the cork passes a brass rod, which is in 
metallic contact with the inner coating, and terminates 
in a ball at the top. 

On presenting the knob of the jar near to the prime 
conductor of an electrical machine, while the latter is 
in operation, a series of sparks passes between the con 
ductor and the jar, which will gradually grow more and 
more feeble, until they cease altogether. The jar is then 
said to be charged. If now we take the discharging-rod, 
which is a curved wire, terminated at each end with a 
knob, and insulated by glass handles (Fig. 234), and apply one 
of the knobs to the outer coating of the jar, and bring the other 
to the knob of the jar, a flash of intense brightness, accompanied 

FIG. 233. 


by a sharp report, immediately ensues. This is the discharge of 
the jar. 

If, instead of the discharging-rod, FIG. 234. 

a person applies one hand to the out 
side of the charged jar, and brings the 
other to the knob, a sudden shock is 
felt, convulsing the arms, and when 
the charge is heavy, causing pain 
through the body. The shock pro 
duced by electricity was first discov 
ered accidentally by persons experi 
menting with a charged phial of water. This occurred in Leyden, 
and led to the construction and name of the Leyden jar. 

418. Theory of the Leyden Jar. This instrument accu 
mulates and condenses great quantities of electricity on its sur 
faces, upon the principle of mutual attraction between unlike 
electricities, one of which is furnished by the machine, the other 
obtained from the earth by induction. First, suppose the outer 
coating insulated; a spark of the positive electricity passes from 
the prime conductor to the inner coating, which tends to repel 
the positive from the outer coating ; but as the latter cannot es 
cape, it remains to prevent, by its counter-repulsion, any addition 
to the charge of the inside, and thus the process stops. But now 
connect the outer coating with the earth, and immediately some 
of its positive electricity, repelled by the charge on the inside, 
passes off, while its negative is attracted close upon the glass, and 
gives room for the accession of more from the earth. The slight 
condensation of negative upon the outside, by its attraction, con 
denses the positive of the inner coating, and allows a second spark 
to pass in from the prime conductor. This produces the same 
effect as the first, and a second addition of negative is made to the 
outer coating, the latter being obtained from the earth as before. 
These actions and reactions go on in a diminishing series, till 
there is a great accumulation of the two electricities, held by mu 
tual attraction as near each other as possible, on opposite sides of 
the glass. The jar in this condition is said to be charged. 

If the positive electricity is on the inner coating, the jar is 
said to be positively charged ; if on the outside, negatively charged. 

419. The Spontaneous Discharge. This occurs when the 
quantities accumulated are so great that their attraction will cause 
them to fly together with a flash and report over the edge of the 
jar. If the glass is soiled or damp, the fluids may pass over and 
mingle with only a hissing noise, in which case it is impossible for 
the jar to be highly charged. 


If the glass is clean and dry, and especially if varnished with 
gum lac, a charge may not wholly disappear for days, or ever 

420. Series of Jars. The same amount of electricity from 
the prime conductor which is required to charge one jar will 
charge an indefinite series, the strength of the charge being less and 
less from the first to the last. This case is analogous to the series 
of conductors (Art. 414). Insulate a series of jars, A, B, C, &c., 
and connect the inner coating of A with the prime conductor, and 
its outer coating with the inner coating of B, the outer of B with 
the inner of C, and so on. Then, as A is charged, the positive 
electricity of its outer coating, instead of passing to the earth, goes 
to the inside of , and that on the outside of B to the inside of 
C, &c., while that on the outside of the last in the series passes to 
the earth. Thus each jar is charged positively by the inductive 
influence of the preceding, just as a series of magnets is formed 
with poles in the same order by a succession of magnetic induc 

421. Division of a Charge in any Given Ratio. If one 

of two jars be charged, and the other not, and if the inner coat 
ings be brought into communication, and also the outer coatings, 
the charge of the first jar is instantly diffused over the two, with 
a report like that of a discharge. In this way a charge may be 
halved, or divided in any other ratio, according to the relative sur 
faces of the jars. 

The self-repellency of each fluid tends to diffuse it over a 
greater surface, and they will be thus diffused if allowed to remain 
within each other s attracting influence; but one of the fluids will 
not be spread over the coatings of another jar, unless opportunity 
is given for both to do it. 

An experiment somewhat resembling the foregoing is this: 
charge two equal jars, one positively, the other negatively, and in 
sulate them both. If the two knobs be connected by a conductor, 
the electricities, notwithstanding their strong attraction, will not 
unite ; for each is held disguised by that on the other side of the 
glass. But if the outer coatings are first connected, then, on join 
ing the knobs, the jars are both discharged at once. 

422. Use of the Coatings. If a jar is made with a wide 
open top, and the coatings movable, then, after charging the jar 
and removing the coatings, very little of the electricities adheres 
to the latter, but nearly the whole remains on the glass. The 
same mutual attraction which condensed them at first still holds 
them there after the coatings are removed. When they come to 


be replaced, the jar can be discharged as usual. But the coatings 
are necessary in charging, to diffuse the electricity over those parts 
of the glass which they cover, and also in discharging, to conduct 
off the whole charge at once. 

423. The Free Portion of an Electrical Charge. Either 
kind of electricity is said to be free when it remains on a body 
only because held by the pressure of the air; but if held by the 
attraction of the opposite kind, it is said to be disguised (Art. 413). 
Nearly all the electricity of a charged jar is disguised, but not the 

The moment after a jar is charged there is a small quantity of 
free electricity on the coating to which the fluid was furnished in 
charging, but not on the other. But after the jar has stood 
charged some minutes, a little is free on both coatings. If the 
charged jar be upon an insulating stand, and the finger brought 
to one coating, a slight spark is taken off; if it be touched again 
immediately, there is no spark, for the free electricity all escaped 
by the first; contact. Let the finger now be brought to the other 
coating, and a spark flies from that. Immediately afterward a 
second spark can be taken from the first coating, and so on alter 
nately for hundreds of times usually before the charge wholly dis 
appears. What is removed at each contact is the free part of the 
charge, which always appears alternately on the two coatings. If 
a small electroscope be connected with each coating, the fluid al 
ternately set free is indicated to the sight. The electroscope on 
the coating which is touched instantly falls, and the other rises. 

424. Explanation of this Phenomenon. The positive 
electricity which is conveyed to the inner coating, in charging a 
jar, attracts to the outer coating from the earth a quantity of the 
negative fluid which is a little less than itself. This is because of 
the thickness of the glass. If it were infinitely thin, the negative 
would be just equal to the positive, and they would neutralize 
each other, and both be perfectly disguised. But as the glass has 
some thickness, the positive exceeds the negative, and disguises it. 
ISTow if the jar, after being charged, is insulated, it is obvious that 
the negative charge on the outer coating cannot disguise all the 
positive (which is more than itself), but only a quantity a little 
less than itself. Hence there must be a little of the positive on 
the inner coating in a free state. By touching the knob, we allow 
this free portion to pass off, and there is left less of the positive in 
the inner coating than there is of the negative in the outer. 
Therefore, all the negative cannot now be disguised, but a slight 
quantity is liberated and ready to pass off as soon as touched. 
And thus, by alternate contacts, the process of discharge goes on, 


the series being longer as the glass is thinner, because then the 
two quantities are more nearly equal. 

425. Electrical Vibrations and Revolutions. If two 

jars be charged in opposite ways, and a figure made of pith be sus 
pended between the knobs by a long thread, it will be attracted by 
that knob whose action on it happens to be greatest. As soon as 
it touches, it is charged with that kind and repelled, and of course 
attracted by the other knob, which is in the opposite state ; thus 
it vibrates between them, causing a very slow discharge of both 
jars. In this case, the outside of the jars not being insulated, the 
electricity, which is slowly set free on the outside, passes off, and 
therefore there is always some free electricity on the knob to be 
imparted to the vibrating figure. 

The electricity of the prime conductor will also cause vibra 
tions, without the use of a jar. Suspend from it a metallic disk 
horizontally a few inches above another which is connected with 
the earth ; then if a glass cylinder surround the two disks so as to 
prevent escape, a number of pith balls between the disks will con 
tinue to vibrate up and down so long as the machine is in action. 
Each ball lying on the lower disk, being electrified by induction 
in the opposite way from the upper one, springs up to it, and 
then, being charged in the same way, is repelled. 

In a similar manner a chime of bells may be rung, orreries re 
volved, &c. 

426. Residuary Charge. If a jar stand charged a few min 
utes, and after the discharge remain some minutes more, then a 
second, and possibly a third, discharge can be made; but these are 
usually very slight. The electricity remaining after the first dis 
charge is called the residuary charge. The larger the jar, and the 
more intense the charge, the larger is this residuum. It is probably 
explained as follows : The charge, at first limited to the coatings, 
gradually diffuses itself on the uncoated glass for a little distance, 
according to the intensity of the charge and the length of time 
the jar remains charged. At the first discharge, only the elec 
tricity which is in contact with the coating is taken off, and that 
which lies on the uncoated glass slowly diffuses itself back again, 
and is conducted over the whole coated surface ; so that, after the 
lapse of a minute or two, a sensible discharge occurs on applying 
the rod a second time. 

427. The Electric Battery. Leyden jars are made of vari 
ous sizes, from a half-pint to one or two gallons. But when a 
great amount of surface is needed, it is more convenient, and, in 
case of fracture by violent discharge, more economical, to connect 



FIG. 235. 

several jars, so that they may be used as one. Four, nine, twelve, 
or even a greater number of jars, are 
set in a box (Fig. 235), whose inte 
rior is lined with tinfoil, so as to con 
nect all the outer coatings together. 
Their inner coatings are also con 
nected, by wires joining all the 
knobs, or by a chain passing round 
all the stems. Care is necessary in 
discharging batteries, that the cir 
cuit is not too short and too perfect, 

since the violence of discharge is liable to perforate the jars. A 
chain, three or four feet long in the circuit, will generally prevent 
the accident. 

428. Different Routes of Discharge. If two or more cir 
cuits are opened at once between the two coatings of a charged 
jar or battery, the discharge will take one or another, or divide be 
tween them, according to circumstances. If the circuits are alike 
except in length, the discharge will follow the shorter. If they 
differ only in conducting quality, the electricities will take the 
lest conductor. If the circuits are interrupted, and in all respects 
alike, except that the conductors of one are pointed at the inter 
ruptions, and of the others not pointed, the discharge will follow 
the line which has pointed conductors. If the circuits are very at 
tenuated (as very line wire, or threads of gold-leaf), the charge is 
liable to divide among them. 

429. Discharging Electrometers. These are instruments 
contrived for measuring the charge in the act of discharging the 
jar. Fig. 236 represents Lane s discharging electrometer. D is a 

FIG. 236. 

FIG. 237. 

rod of solid glass, which holds the metallic rod and balls B C. 
This rod, being in a horizontal position at the height of the knob 
A, can be placed at any desired distance from it, Then, if the 
circuit through which the charge is to be sent extends from the 


rod B C to the external coating, the interval of air between 4 and 
B is all which prevents discharge ; and as soon as the charge is 
increased, till its tension is sufficient to leap that interval, the dis 
charge will take place. The greater that space is, of course the 
greater the charge must be, before it will pass across. If A and B 
are in contact, no charge at all will collect. 

The unit jar is used to measure the charge of another jar, by 
conveying to it successive equal charges of its own. A B (Fig. 
237) is the instrument, consisting of a small open jar, placed hori 
zontally on an insulating stand, B. From the metallic part of 
the support, the bent rod and ball, C, come near to D, the rod of 
the inner coating, and can be turned so as to increase or diminish 
its distance. Let the knob, />, be near the prime conductor, and 
that of the outer coating near the top of the jar, E, which is to be 
charged, the outside of the latter being in communication with 
the earth. While A is charging, the positive electricity of its 
outer coating goes to the inner coating of the large jar, and par 
tially charges it. Presently the unit jar discharges spontaneously 
across the distance between D and C. It is then in the neutral 
condition, as at first, and the process is repeated till E is charged 
with the requisite number of units. 

430. The Effect of a Point Presented to an Electri 
fied Body. It has been noticed (Art. 406) that a pointed wire 
attached to the prime conductor wastes the charge very quickly, 
because of the accumulation at the point. The conductor loses 
its charge just as quickly by presenting a pointed rod toward it. 
For the induced electricity of the rod and person holding it is in 
like manner accumulated at the point, and readily escapes to 
mingle with and neutralize its opposite in the prime conductor. 
Thus the charge disappears at once. In a similar way is to be ex 
plained the use of the points on the prime conductor presented to 
the glass plate. When the two electricities are separated at the 
surface of contact between the plate and rubbers, the plate is posi 
tively electrified. This positive charge acts inductively on the 
prime conductor, attracting the negative kind to the points, where 
it passes off and neutralizes what is on the plate, and leaves a pos 
itive charge on the prime conductor. 

431. The Gold-Leaf Electroscope. The principle of in 
duction explains the construction of some other instruments be 
sides the Leyden jar ; as the gold-leaf electroscope, the electrical 
condensers, and the electrophorus. The first has been already 
described (Art. 387) ; we have only to explain its operation by the 
principle of induction. Let a body positively electriiied be 
brought within a few feet of the knob. It attracts the negative 



FIG. 238. 

from the leaves into the knob, and repels the positive from the 
knob into the leaves ; they are thus electrified alike, and repel 
each other. If the charged body is brought so near that the leaves 
touch the conductors, which are placed on the sides of the cylin 
der, and discharge their induced electricity to them, then they 
collapse. After this, they will diverge again, whether the electri 
fied body is brought still nearer, or withdrawn ; if brought nearer, 
they diverge by means of a new portion of positive, repelled from 
the knob; if withdrawn, they diverge by the return of negative 
electricity from the knob, which is no longer neutralized by the 
positive, since the latter has been discharged to the earth. 

432. The Electrical Condenser. Instruments called by 
this name are intended for the accumulation of electricity from 
some feeble source, until it may be rendered sensible. The most 
delicate is the gold-leaf condenser. Suppose the gold-leaf electro 
scope to have a disk, A, instead of a knob on the top (Fig. 238). 
Another disk, B, is furnished with an insulating handle, 

and between the disks is placed the thinnest possible 
non-conductor, as a film of Tarnish. Bring the finger 
in contact with the under-side of A, to connect it with 
the earth. Then bring to the upper side of #the source 
of feeble electricity (as, for example, a piece of copper, 
after being touched to a piece of zinc), the small quan 
tity of electricity imparted to B induces an equal 
amount of the opposite in A, drawn in from the earth. 
After the disks have touched each other again, a second 
contact upon B repeats the action ; and when this has 
been done a great number of times, there are condensed on the two 
sides of the varnish small charges which are held in that state by 
induction. As yet, the gold-leaves are at rest ; but on removing 
the finger from A, and taking up B by the insulating handle, the 
electricity condensed in A is set free, flows down to the leaves and 
repels them, thus rendering the accumulation perceptible. 

433. The Electrophorus. This is a very simple electrical 
machine for giving the spark. It consists FlG 339 

of a circular cake of resin in a wooden 
base, A (Fig. 239), and a metallic disk, B 9 
having a glass handle. Excite the resin by 
fur or flannel ; set the disk B upon it, and 
touch the latter with the finger. The disk 
now has a disguised charge of positive elec 
tricity, drawn in by the negative charge of 
the plate through the finger. On lifting 
the disk by the handle, its charge is set free, and may be taken 



off in a brilliant spark. Set the disk down again, touch it, and 
lift it, and the same thing occurs, even hundreds or thousands of 
times, and, after standing for hours, is ready to operate still in 
the same way. 

This case of inductive action seems at first perplexing, because 
there is no glass plate, no film of varnish, no non-conductor of 
any kind, between the two opposite electricities of the resin and 
disk. "Why then do they not at once mingle, and neutralize each 
other ? It is simply because the resin is an excited body, the neg 
ative electricity having been developed upon it by friction. When 
we touch the finger to the disk, the positive that enters does meet 
the negative, and neutralize it for the time being ; but on separa 
ting the plate and disk, the electricities also separate, as is always 
the case when an electric and the rubber are removed from each 
other after friction. Thus one charge on the resin may be made 
to induce any number of successive charges upon the disk. 

434. The Induction Machine. This instrument (Fig. 240), 
known also as the Holtz machine, from the name of the inventor, 

FIG. 240. 

develops electricity with great rapidity without friction, except as 
it is employed for a moment at first to electrify one of the sectors, by 
the side of which the plate revolves. This electrified sector acts in 
ductively on the successive portions of the revolving plate, with 
out losing sensibly its own electric charge, just as the electi opho- 


rus plate induces charge after charge on the metallic disk, without 
loss to itself.* 

435. The Leichtenberg Figures. When a spark of elec 
tricity is laid upon a non-conductor, it will, by its own self-repel- 
lency, extend itself a little distance along the surface. The Leich- 
teiiberg figures furnish a visible illustration of this fact, and also 
show that the two fluids diffuse themselves in very different forms. 
Lay down sparks of positive electricity from the knob of the Ley- 
den jar upon a plate of resin, and near them some sparks of nega 
tive electricity. Then blow upon the plate the mingled powders 
of sulphur and red-lead. The sulphur, by the agitation of passing 
through the air, will be electrified negatively, and attracted there 
fore by the positive sparks ; the red-lead, positively electrified, will 
be attracted by the negative. Thus the spots on which the elec 
tricities are placed will appear in their exact forms by means of 
the colored powders attached to them. The positive resemble 
stars, or rather a group of crystals shooting out from a nucleus; 
the negative spots are circles with smooth edges ; and the size of 
the electrified spots in each case depends on the quantity of elec 
tricity in the spark. 



436. Variety of Effects. Some of the effects of electrical 
discharges have been incidentally noticed in the foregoing chap 
ters. The bright light, the sharp sound, and the great suddenness 
of the transmission, are remarkable phenomena in every discharge 
of a Ley den jar or battery. The various effects may be classified 
as luminous, mechanical, chemical, and physiological. 

437. Luminous Effects. Light is seen only when elec 
tricity is discharged in considerable quantities through an ob 
structing medium. Hence, no light is perceived when it flows 
through a good conductor, unless of very small diameter. But if 
there is the least interruption, or if the conductor is reduced to a 
very slender form, then light appears at the interruption, and at 
those parts which are too small to convey the electricity. Thus, 

* The Holtz machine has been greatly improved by Mr. E. S. Ritchie, of 
Boston. The figure presents this improved form. Physicists are not fully 
agreed r as to the mode of explaining all the phenomena of this machine. 


the discharge of a battery through a chain gives a brilliant scintil 
lation at every point of contact between the links, 

438. Modifications of the Light The length, color, and 
form of the electric spark vary with the nature and form of the 
conductors between which it passes, and with the quality of the 
medium interposed between them. 

Electrical sparks are more brilliant in proportion as the sub 
stances between which they occur are better conductors. A spark 
received from the prime conductor upon a large metallic ball is 
short, straight, and white; on a small ball it is longer, and 
crooked ; received on the knuckle, a less perfect conductor, the 
middle part is purplish ; on wood, ice, a wet plant, or water, it is 

From a point positively electrified, the electricity passes in the 
form of a faint brush or pencil of rays ; a point connected with 
the negative side exhibits a luminous star. 

When electricity passes through rarefied air, the light becomes 
faint, and is generally changed in color. The electrical spark, which 
in common air is interrupted, narrow, and white, becomes, as the 
rarefaction proceeds, continuous, diffused, and of a violet color, 
which tint it retains as long as it can be seen. If a battery is dis* 
charged through a tube several feet long, nearly exhausted of air, 
the whole space is filled with a rich purple light. The sparks 
from the machine, conveyed through the same tube, exhibit flash 
ings and tints exceedingly resembling the Aurora Borealis. 

The Geissler tubes are tubes of complex forms, and containing 
a slight trace of some gas or vapor, which show various colors 
and intensities of electric light, according to the kind of gas, the 
diameter of the parts, and the quality of the glass. The electricity 
is conveyed into the tubes by platinum wires sealed into their ex 

Various colors are obtained by sending charges through differ 
ent substances. An egg is bright crimson ; the pith of cornstalk, 
orange ; fluor-spar, green ; and loaf-sugar, white and phosphor 

439. Luminous Figures. Metallic conductors, if of suffi 
cient size, transmit electricity without any luminous appearance, 
provided they are perfectly continuous ; but if they are separated 
in the slightest degree, a spark will occur at every separation. 
On this principle, various devices are formed, by pasting a narrow 
band of tinfoil on glass, in the required form, and cutting it across 
with a penknife, where we wish sparks to appear. If an inter 
rupted conductor of this kind be pasted round a glass tube in a 
spiral direction, and one end of the tube be held in the hand, and 



FIG. 241. 

the other be presented to an electrified conductor, a coil of bril 
liant points surrounds the tube. Words, flowers, and other com 
plicated forms, are also produced nearly in the same manner, by a 
suitable arrangement of interruptions in a narrow line of tinfoil, 
running back and forth on a plate of glass. 

4iO. Mechanical Effects. Powerful electric discharges 
through imperfect conductors produce certain mechanical effects, 
such as perforating, tearing, or breaking in pieces, which are all 
due to the sudden and violent repulsion between the electrified 

A discharge through the air is supposed to perforate it. If 
the air through which the spark is passed lies partially inclosed 
between two bodies which are easily moved, the force by which 
the air is rent will drive them asunder. Thus, a little block may 
be driven out from the foundation of a 
miniature building, and the whole be top 
pled down. But this enlargement of in 
closed air is best seen in Kinnersley s air 
thermometer (Fig. 241). As the spark 
passes between the knobs in the large 
tube, the air confined in it is suddenly 
driven asunder, so as to press the water 
which occupies the lower part two or 
three inches up the tube, as represented. 
As soon as the discharge has occurred, 
the water quietly returns to its level. The 
sharp sound which is produced by the dis 
charge of a Ley den jar is due to the sud 
den compression of the air, and also to the 
collapse which immediately succeeds. 

The path of the electric spark through 
the air, when short, is straight; but if 
more than about four inches long, is usually crinkled. This is 
supposed to arise from the condensation of the air before it, by 
which it is continually turned aside. 

When the charge is passed through a thick card, or the cover 
of a book, a hole is torn through it, which presents the rough ap 
pearance of a bur on each side. By means of the battery, a quire of 
strong paper may be perforated in the same manner; and such is 
the velocity with which the fluid moves, that if the paper be freely 
suspended, not the least motion is communicated to it. Pieces of 
hard wood, of loaf-sugar, and brittle mineral substances, are split 
in two, or shivered to pieces, by an intense charge of a battery. 
But good conductors of much breadth are not thus affected. The 


charge, as it is transmitted, passes over the whole body, instead of 
being concentrated in any one line. But if liquids which are good 
conductors are closely confined on every side, they show that a, 
violent expansion is produced by a discharge. Thus, when a 
charge is sent through water confined in a small glass tube or 
bull, the glass is shattered to pieces ; and mercury in a thick cap 
illary tube is expanded with a force sufficient to splinter the glass. 

441. Chemical Effects. These are various : combustion of 
inflammable bodies ; oxydation, fusion, and combustion of metals; 
separation of compounds into their elements; reunion of elements 
into compounds. 

Ether and alcohol may be inflamed by passing the electric 
spark through them ; phosphorus, resin, and other solid combus 
tible bodies, may be set on fire by the same means ; gunpowder 
and the fulminating powders may be exploded, and a candle may 
be lighted. Gold-leaf and fine iron wire may be burned, by a 
charge from the battery. Wires of lead, tin, zinc, copper, plati 
num, silver, and gold, when subjected to the charge of a very large 
battery, are burned, and converted into oxides. 

The same agent is also capable of restoring these oxides to 
their simple forms. Water is decomposed into its gaseous ele 
ments, and these elements may again be reunited to form water. 
By passing a great number of electric charges through a confined 
portion of air, the oxygen and nitrogen are converted into nitric 
acid. The ozone which is almost always perceived in connection 
with electrical experiments is to be considered as one of the chem 
ical effects of electricity. 

Galvanic electricity is a form of this agent much better adapted 
than frictional electricity to produce chemical as well as magnetic 

442. Physiological Effects. The shock experienced by the 
animal system, when the charge of a jar passes through it, has 
been already mentioned. 

A slight charge of the Leyden jar, passed through the body 
from one hand to the other, affects only the fingers or the wrists; 
a stronger charge convulses the large muscles of the arms; a still 
greater charge is felt in the breast, and becomes somewhat pain 
ful. The charge of a large battery is sufficient to destroy life, if it 
lw sent through the vital organs. By connecting the chains 
which are attached to the jar with insulating handles, it is easy to 
pass shocks through any particular joint, muscle, or other part of 
the body, as is frequently done for medical purposes. 

The charge may be passed through a great number of persons 
at the same time. Hundreds of individuals, by joining hands, 


have received the shock at once, though there is more difficulty in 
passing a charge of given intensity as the number is increased. 

If the spark is taken by a person from the prime conductor, 
the quantity is not sufficient, unless the conductor is of extraordi 
nary size, to produce what is called the shock ; a pricking sensa 
tion in the flesh where the spark strikes, and a slight spasm of the 
muscle, is all that is noticeable. A person may make his own 
body a part of the prime conductor by standing on an insulating 
stool that is, a stool having glass legs, and touching the conduc 
tor of the machine. This occasions no sensation at all, except 
what arises from the movement of the hair, in yielding to the re- 
pellency of the fluid. If another person takes the spark from him, 
the prick is more pungent, as the quantity is larger than in the 
prime conductor alone. 

443. Velocity of Electricity. This is so great that no ap 
preciable time is occupied in any case of discharge. When we 
seem to see lightning move from the cloud to the earth, we find 
that such a progress is imagined, not perceived ; for, by a little 
effort, we can just as well learn to see it pass from the earth to the 

Wheatstone a few years since devised an ingenious method of 
measuring the time in which electricity passes over a wire only 
half a mile long. The wire was so arranged that three interrup 
tions, one near each end, and one in the middle of the wire, were 
brought side by side. When the discharge of a jar was transmitted, 
the sparks at these interruptions were seen by reflection in a 
swiftly revolving mirror. An exceedingly small difference of 
time between the passage of those interruptions could be easily 
perceived by the dis2}lacement of the sparks as seen in the whirling 
mirror. The amount of observed displacement and the known 
rate of revolution of the mirror, would furnish the interval of time 
occupied by the electricity in passing from one interruption to the 
next. By a series of experiments, Wheatstone arrived at the con 
clusion that, on copper wire, one-fifteenth of an inch in diameter,, 
electricity moves at the rate of 288,000 miles per second, a velocity 
much greater than that of light. 

Galvanic electricity moves very much slower. Its rate on iron 
wire, of the size usually employed for telegraph lines, is about 
16,000 miles per second. 




444. Electricity in the Air. The atmosphere is always 
more or less electrified, sometimes pos tively, sometimes negatively. 
This fact is ascertained by several different forms of apparatus. 
For the lower strata, it is sufficient to elevate a metallic rod a few 
feet in length, pointed at the top, and insulated at the bottom. 
With the lower extremity is connected an electroscope, which in 
dicates the presence and intensity of the electricity. For experi 
ments on the electricity of higher portions, a kite is employed, 
with the string of which is intertwined a fine metallic wire. The 
lower end of the string is insulated by fastening it to a support of 
glass, or by a cord of silk. If a cloud is near the kite, the quan 
tity of electricity conveyed by the string may be greatly increased, 
and even become dangerous. Cavallo received a large number of 
severe shocks in handling the kite-string ; and Richman, of Peters- 
burgh, was killed by a discharge of electricity which came down 
the rod which he had arranged for his experiments, but which was 
not provided with a conductor near by it, for taking off extra 

The electricity of the atmosphere is most developed when hot 
dry weather succeeds a series of rainy days, or the reverse ; and 
during a single day, the air is most electrical when dew is begin 
ning to form before sunset, or when it begins to exhale after sun 
rise. In clear, steady weather, the electricity is generally positive; 
but in falling or stormy weather, it is frequently changing from 
positive to negative, and from negative to positive. 

445. Thunder-Storms. Thunder-clouds are, of all atmos 
pheric bodies, the most highly charged with electricity ; but all 
single, detached, or insulated clouds, are electrified in greater 
or less degrees, sometimes positively and sometimes negatively. 
"When, however, the sky is completely overcast with a uniform 
stratum of clouds, the electricity is much feebler than in the single 
detached masses before mentioned. And, since fogs are only 
clouds near the surface of the earth, they are subject to the same 
conditions : a driving fog, of limited extent, is often highly elec 

Thunder-storms occur chiefly in the hottest season of the year, 
and after mid-day, and are more frequent and violent in warm 
than in cold countries. They never occur beyond 75 of latitude 


seldom beyond C5. In the New England States they usually 
come from the west, or some westerly quarter. 

The storm itself, including everything except the electrical ap 
pearances, is supposed to be produced in the same manner as 
other storms of wind and rain ; and the electricity is developed by 
the rapid condensation of watery vapor, and by friction. Elec 
tricity is not to be regarded as the cause, but as a consequence or 
concomitant of the storm. But the precipitation of vapor must be 
sudden and copious, since when the process is slow, too much of 
the electricity evolved would escape to allow of the requisite accu 
mulation. Also, if a storm-cloud is of great extent, it is not likely 
to be highly electrified, because the opposite electricities, which 
may be developed in different parts of it, have opportunity to 
mingle and neutralize ; and points of communication with the 
earth will here and there occur. Clouds of rapid formation, violent 
motion, and limited extent, are therefore most likely to be thun 

446. Lightning. When a cloud is highly charged, it operates 
inductively on other bodies near it, such as other clouds, or the 
earth. Hence, discharges will occur between them. Lightning 
passes frequently between two clouds, or even between two parts 
of the same cloud, in which opposite electricities are so rapidly de 
veloped that they cannot mingle by conduction. But, in general, 
the discharges of lightning take place between the electrified cloud 
and the earth, whose nearer part is thrown into the opposite elec 
trical state by induction. It is supposed that, in some instances, 
a discharge occurs between two distant clouds by means of the 
earth, which constitutes an interrupted circuit between them. 
The crinkled form of the path of lightning is explained in the 
same way as that of the spark from the machine, and the thunder 
is caused by the simultaneous rupture and collapse of air in all 
parts of the line of discharge. The words chain-lightning, sheet- 
lightning, and heat-lightning, are supposed not to indicate any real 
differences in the lightning itself, but only in the circumstances 
of the person who observes it. If the crinkled line of discharge is 
seen, it is chain or fork lightning ; if only the light which pro 
ceeds from it is noticed, it is s/^-lightning; if, in the evening, 
the thunder-storm is so far distant that the cloud cannot be seen, 
nor the thunder heard, but only the light of its discharges can be 
discerned in the horizon, it is frequently called fofltf-lightning. 

447. Identity of Lightning and Electrical Discharges. 

Franklin was the first to point out the resemblances between the 
phenomena of lightning and those of frictional electricity. He 
was also the first to propose the performance of electrical experi- 


ments by means of electricity drawn from the clouds. The points 
of resemblance named by Franklin were these: 1. The crinkled 
form of the path. 2. Both take the most prominent points. 
3. Both follow the same materials as conductors. 4. Both inflame 
combustible substances. 5. They melt metals in attenuated forms. 
6. They fracture brittle bodies. 7. Both have produced blindness. 
8. Both destroy animal life. 9. Both affect the magnetic needle in 
the same manner. In 1752, he obtained electricity from a thunder 
cloud by a kite, and charged jars with it, and performed the usual 
electrical experiments. 

418. Lightning-Rods. Franklin had no sooner satisfied him 
self of the identity of electricity and lightning than, with his usual 
sagacity, he conceived the idea of applying the knowledge acquired 
of the properties of the electric fluid so as to provide against the 
dangers of thunder-storms. The conducting power of metals, and 
the influence of pointed bodies, to transmit the fluid, naturally 
suggested the structure of the lightning-rod. The experiment 
was tried, and has proved completely successful; and probably no 
single application of scientific knowledge ever secured more celeb 
rity to its author. 

Lightning-rods are often constructed of wrought iron, about 
three-fourths of an inch in diameter. The parts may be made 
separate, but, when the rod is in its place, they should be joined 
together so as to fit closely, and to make a continuous surface, 
since the fluid experiences much resistance in passing through 
links and other interrupted joints. At the bottom the rod should 
terminate in two or three branches, going off in a direction from 
the building, and descending to such a depth that they will reach 
permanent moisture. At top the rod should be several feet higher 
than the highest parts of the building. It is best, when practica 
ble, to attach it to the chimney, which needs peculiar protection, 
both on account of its prominence and because the products of the 
combustion, smoke, watery vapor, &c., are conductors of elec 
tricity. For a similar reason, a kitchen chimney, being that in 
which the fire is kept during the season of thunder-storms, re 
quires to be especially protected. The rod is terminated above in 
one or more sharp points ; and as these points are liable to lose 
their sharpness, and have their conducting power impaired by 
rust, they are protected from corrosion by being covered with 
gold-leaf or silver-plate. Rods may be made of smaller size than 
above described; but if so, there should be a proportionally 
greater number. It is well to connect with the rods, and with the 
earth, all extended conductors upon or within the building, such 
as metallic coverings of roofs, water conductors, bundles of bell- 


wires, &c. ; in order that large discharges may have opportunity 
to divide, and take several circuits, without doing injury at the 
non-conducting intervals. 

449. In what way Lightning-Rods Afford Protection. 
Lightning-rods are of service, not so much in receiving a discharge 
when it conies, as in diminishing the number of discharges in their 
vicinity. They continually carry on a silent communication be 
tween the two electricities, which are attracting each other, one 
in the cloud, the other in the earth ; so that a village well fur 
nished with rods has few discharges of lightning in it. All tall 
pointed objects, like spires of churches and masts of ships, exert a 
similar influence, though in a less degree, because not so good 

During a thunder-storm, or immediately after it, if a person 
can be near the top of a high rod, he will sometimes hear the hiss 
ing sound of electricity escaping from it, as from a point attached 
to the prime conductor of a machine. In the same circumstances, 
if it were quite dark, he would probably see the brush or star of 
light on the point. The statement of Caesar in his Commentaries, 
" that the points of the soldiers darts shone with light in the 
night of a severe storm," probably refers to the visible escape of 
electricity from the weapons as from lightning-rods. 

450. Protection of the Person. Silk dresses are some 
times worn with the view of protection, by means of the insula 
tion they afford. They cannot, however, be deemed effectual un 
less they completely envelop the person ; for if the head and the 
extremities of the limbs are exposed, they will furnish so many 
avenues as to render the insulation of the other parts of the sys 
tem of little avail. The same remark applies to the supposed se 
curity that is obtained by sleeping on a feather bed. Were the 
person situated ivitliin the bed, so as to be entirely enveloped by 
the feathers, they would afford some protection ; but if the person 
be extended on the surface of the bed, in the usual posture, with 
the head and feet nearly in contact with the bedstead, he would 
rather lose than gain by the non-conducting properties of the bed, 
since, being a better conductor than the bed, the charge would 
pass through him in preference to that. If the bedstead were of 
iron, its conducting quality would probably be a better protection 
than the insulating property of the feathers, since, by taking the 
charge itself, it would keep it away from the person. So, a man s 
garments soaked with rain have been known to save his life, being 
a better conductor than his body. Animals under trees are pecu 
liarly exposed, because the trees by their prominence are liable to 
be the channels of communication for the electric discharge, and 


the animal body, so far as it reaches, is a better conductor than 
tho tree. Tall trees, however, situated near a dwelling-house, fur 
nish a partial protection to the building, being both better con 
ductors than the materials of the house, and having the advantage 
of superior elevation. 

451. How Lightning Causes Damage. The word strike, 
which is used with reference to lightning, conveys no correct idea 
of the nature of the movement of electricit}*, or of the injury 
which it causes. One kind of electricity, developed in a cloud, 
causes the other to be accumulated by induction in the part of the 
earth nearest to it. These electricities strongly attract each other ; 
consequently, that in the earth presses upward into all prominent 
conducting bodies toward the other ; and, if those bodies are nu 
merous, high, the best of conductors, and terminated by points, 
the electricity will flow off from them abundantly, and mingle 
with its opposite in the air above ; and thus discharges are in a 
great degree prevented. But if these channels for silent commu 
nication are not furnished, the quantity of electricity will increase, 
till the strength of attraction becomes so great that the fluid will 
break its way through the air, usually from some prominent ob 
ject, as a building or tree, and thus the union of the two elec 
tricities takes place. The building or tree in this case is said to 
b3 struck ly lightning ; it is rent, or otherwise injured, by the 
great quantity of electricity which passes violently through it, in 
an inconceivably short space of time. The effects produced are 
exactly like those caused by discharges of the electrical battery, 
on a greatly enlarged scale. The charge of a large battery, taken 
through the body in the usual way, would prostrate a person by 
the violence of the shock ; but the same charge, if allowed to oc 
cupy a few seconds in passing by means of a point, would not be 
felt at all. 

Fulgurites are tubes of silicious matter formed in the ground, 
where lightning has struck in sandy soil, and melted the sand 
around its path. 





452. Electricity Developed by Chemical Action. In 

a glass vessel (Fig. 242) containing a mixture of one part of sul 
phuric acid and seven or eight parts of 
water, put two plates, one of copper, C, 
and the other of zinc, Z, to each of which 
is soldered a copper wire. On bringing 
the extreme ends of the wires together, a 
feeble flow of electricity will take place 
through the wires, the plates, and the 
liquid. This is called the galvanic or vol 
taic current of electricity. It is developed 
by the chemical action of the acid on the 
metals; and this condition of electricity 
is called galvanic or voltaic, from Galvani and Volta, two Italian 
philosophers, who made the first discoveries of importance in this 
branch of science. It is also called dynamical electricity, for rea 
sons to be mentioned hereafter. 

453. Definitions. An element or cell is a jar containing any 
arrangement of substances for the purpose of obtaining the gal 
vanic electricity. A battery is a number of elements properly con 
nected with each other. 

The poles or electrodes of a cell or battery are the extremities 
of the wires where the electricities appear. 

The circuit is the path or conductor provided for the flow of 
the current that is, the liquid, the plates, and the wires. The 
circuit is said to be closed when the wires are joined, so that there 
is a flow of the current; when they are separated, the current 
ceases, and the circuit is said to be broken, or to be open. 


454. The Essential Parts of an Element. An clement 
must consist of two unlike substances (they are generally tAvo dif 
ferent metals), separated ly continuous moisture. 

Volta s original battery, called the dry pile, consisting of alter 
nate disks of copper, zinc, and paper, was no real exception, since 
the paper absorbed sufficient moisture from the atmosphere. 

455. The Cell of Two Fluids. An element of copper, 
zinc, and dilute acid, already described, soon loses its efficiency. 
Improved batteries, by which a constant flow of electricity may be 
maintained for a considerable length of time, are those in which 
two liquids are employed, and generally some other substance than 
copper for one of the metals. The liquids must be separated by 
some porous substance, which shall prevent them from mingling, 
and at the same time, being saturated by the liquid, shall not in 
terrupt the necessary moist communication between the metals. 

456. Constant Batteries. Batteries composed of cells con 
taining two liquids arc called constant, because their action con 
tinues for so long a time without sensible abatement. Among the 
best of these is Grove s battery, one element of which is shown in 
Fig. 243, which represents a glass jar contain 
ing a hollow cylinder of zinc, which has a 

narrow opening on one side from top to bot 
tom, that the liquid in which it is placed may 
circulate freely within it. Within the zinc is 
a cylindrical cup of porous earthenware, and 
within that is suspended a lamina of plati 
num. One of the circuit wires is in metallic 
communication with the zinc, and the other 
with the platinum, by means of the binding 
screws at the top. The earthen cup is now 
filled with strong nitric acid, while the space outside of it, in 
which the zinc is placed, contains dilute sulphuric acid. It is 
necessary to amalgamate the surface of the zinc with mercury, in 
order to prevent the action of the acid when the circuit is broken. 
Bunsen s lattery is the same as Grove s, except that in it a 
cylinder of carbon is used instead of a leaf of platinum, on account 
of the expense of the latter. It is very generally employed in 
telegraphy. Fig. 244 is a Bunsen battery of ten cells. 

457. Direction of the Current. Both positive and nega 
tive electricities are furnished by a galvanic battery. In one of 
copper and zinc, the former is found at the extremity of the wir3 
connected with the copper plate, which extremity is therefor 1 
called the positive electrode. For a corresponding reason, the 


other electrode is the negative one. On the supposition that elec 
tricity is a fluid (a hypothesis which is now discarded, though the 
convenient terms which it gave rise to, as current, floiv, &c., are 

FIG. 244. 

retained), there are manifestly two currents, flowing in opposite 
directions. For the sake of convenience, only the positive one is 
spoken of as the current. The direction in which this passes 
through the wires is from the copper to the zinc. 

458. Galvanio and Frictional Electricity Compared. 

The electricities furnished by chemical action and by friction are 
undoubtedly the same in kind. But they differ m that the former 
i.3 produced in greater quantity, while the latter is in a state of 
greater intensity, or tension. This will be understood by referring 
to heat. The quantity of heat in a warm room is vastly greater 
than that in the flame of a lamp ; yet the former is agreeable, 
while the latter, if touched, causes severe pain by its greater inten 
sity. In a similar manner, a quantity of galvanic electricity may 
pass through the body without harm, which, if it possessed the in 
tensity of frictional electricity, would instantly destroy life. 

The word tension, or intensity, expresses the degree of force ex 
erted by electricity in overcoming a given obstacle, as a break in a 

(1) From this difference in quantity and intensity results a 
very great difference in continuance of action. This is indicated by 
the terms dynamical and statical. Galvanic electricity, being pro 
duced in prodigious quantities and with very feeble tension, may 
flow in a steady, gentle stream for many hours, and is hence called 
dynamical. While frictional electricity, being small in quantity and 
intense in action, darts through an opposing medium instantane 
ously, and with great violence. What motion it has is therefore 


merely incidental to its passage from one state of rest to another. 
Hence the propriety of the term statical. 

(2) Again, owing to its low tension, galvanic electricity will 
traverse many thousands of feet of wire rather than pass through 
the thin covering of silk with which the wire is insulated, and 
which would be but a slight obstacle in the path of frictional 

(3) Analogous to the latter is its inability to pass from one 
conductor to another in its immediate vicinity. In order to es 
tablish the flow of a current, the electrodes must first be brought 
into actual contact, or exceedingly near to each other. They may 
then be separated more or less, according to the intensity of the 
battery, without interrupting the current. 

459. Actual Amount Comparisons have been made of the 
actual quantities of electricity obtained by chemical action and by 
friction. Faraday has shown that to decompose one grain of 
water into its constituent elements, oxygen and hydrogen, requires 
an amount of frictional electricity equal to the charge of a Leyden 
battery with a metallic surface of thirty -two acres, equal to a very 
powerful flash of lightning. But by a galvanic current, the same 
result is accomplished in three minutes and forty-five seconds. 
From this some idea may be formed of the vast quantity of elec 
tricity produced during the steady flow for several hours of a 
Grove or Bunseu. battery. 

460. Quantity and Tension Regulated. It may be stated 
in general that quantity increases with the surface of metal, and 
intensity with number of elements. Thus, from an element which 
presents two square feet of surface of metal to the action of the 
acids, we obtain a greater quantity of electricity than from one 
whose metallic surface is one square foot, but no increase of ten 
sion. On the other hand, fi^m two elements, each of one square 
foot of surface, we find greater tension, but no increase in quantity. 

461. Manner of Connecting the Elements of a Bat 
tery. AVhen quantity of electricity is desired, all the plates of 
the same name in the several elements should be united by con 
necting wires, as, for example, all the zinc plates together, and all 
the copper together. The battery thus becomes substantially a 
single large cell, and is called a quantity lattery. 

When tension is sought for, the zinc of one cell should be 
joined to the copper of the next, and so on through the series. A 
battery thus formed is called an intensity battery. 

462. Effects. The presence of a galvanic current is indicated 
by certain chemical, physical, physiological, or magnetic effects. 


An example of the first is the decomposition of water, already 

A physical effect is the production of light. When the elec 
trodes are brought together, and then separated, a spark is pro 
duced of varying intensity and duration. 

The shock which is felt when the electrodes are held in the 
hands, and which affects more or less of the person, is a physio 
logical effect. 

The magnetic properties of a current will be spoken of here 

463. Size of Battery for Required Results. The phys 
ical or physiological results obtained from a single element of or 
dinary size of any kind are quite limited. No shock can be ob 
tained from the direct current of a single cell. But a smart one is 
given by fifty Bunsen cells. It is felt only at the instant of closing 
or breaking the circuit. A shock from a battery of several hundred 
cells would affect the system painfully, if not dangerously. Nine 
hundred cells of copper, zinc, and dilute acid, furnish an arch of 
flame between the electrodes six inches in length. Brilliant re 
sults are also obtained from twenty Grove cells. 

Such magnetic and chemical results as require a current of low 
intensity and small quantity may readily be obtained from a single 
cell. Such are electro typing or the deflection of the magnetic 


E L E C T R 0-M A G N E T I S M. 

464. Helices. A wire bent in a spiral, as in Fig. 245, is 
called a coil or helix. If the wire is coiled in the direction of the 
thread of a common or 
right-hand screw (Art. 136), FlG - 245 

it is called a right-hand 
helix; if in the direction 
of the thread of a left-hand 
screw, it is called a left-hand FlG - 246 - 

helix. Without referring to 
the screw, the distinction 
between the right and left 
hand helix may be described thus: When a person looks at a 
helix in the direction of its length, if the wire, as it is traced from 


him, winds from the left over to the right, it is a right-hand helix 
(Fig. 245) ; if from the right over to the left, a left-hand helix 
(Fig. 246). 

465. The Solenoid. Let a helix be constructed as in Fig. 
247, in which the ends are turned back through the coil, metallic 
contact being avoided through 
out; this is called a solenoid 
that is, a tubular or channel- 
shaped magnet. Next, let the 
electrodes p and n of a battery 
be furnished with sockets, one 
vertically above the other, in 
which the two ends of the helix 
wire are placed. The solenoid 
is then free to turn nearly a whole revolution around a vertical 
axis, at the same time that a current is passing through it. The 
helix is supposed to be a left-hand one, and is so connected with 
the battery that the current passes through it from N to S, and 
therefore around it from right over to left. 

While the current flows, the following phenomena may be ob 
served : 

1. If a magnet be brought near it, N will be attracted by the 
south pole, and S by the north pole. If, instead of a magnet, 
another solenoid be presented to it, whose corresponding extremi 
ties are N and S , N and S will attract each other, as also S 

2. If not disturbed, the coil will place itself lengthwise in the 
direction of the magnetic meridian, with the extremity N toward 
the north, and S toward the south. 

3. If a bar of iron be placed within it, the bar will become a 
magnet, having its north pole at JV, and its south pole at S. 

If a right-hand helix had been employed, all these phenom 
ena would have been reversed. 

466. Ampere s Theory of Magnetism. In these experi 
ments a coil is found to act the same as a magnet whose north 
and south poles are at N and S respectively. We therefore de 
duce the following: 

1. A helix traversed by a galvanic current is a magnet the 
position of whose poles depends on the direction of the current. 

2. Conversely, a magnet, like a coil, may be conceived to owe 
its magnetic properties to currents of electricity which traverse it. 

This is the theory of Ampere, and is the one generally received, 
notwithstanding some objections to it. 

In the helix a single current is present. But in a magnet we 



must conceive of an infinite number of currents, the circuit of 
each being confined to an individual molecule. Fig. 248 repre 
sents a magnet accord 
ing to this theory, and Fia - R 
N and 8 (Fig. 249) 
show the extremities 
of the north and south 

poles on a larger scale. The arrows on the convex surface show 
the general direction of all the currents that is, of those portions 
cf them nearest the surface, 
where magnetism is in fact 
developed and may there 
fore represent them all. 

Since S is the south pole 
of the magnet, as supposed 
to be seen by an observer 
looking at it in the direc 
tion of its axis, it follows 
that when a magnet is in 
its normal position, that 
is, with its north pole point 
ing northward, its currents circulate from west over to east, and 
therefore from left over to right if the observer is also looking 
northward. In like manner, it is evident that to a person looking 
along the length of a magnet, from its north toward its south pole, 
the currents circulate from the right over to the left. 

These supposed currents of the magnet are so small that we 
cannot take cognizance of them directly. But on the basis of 
Ampere s theory, we may substitute for them the large and man 
ageable current of a helix. Then, by determining experimentally 
the causes of magnetic phenomena in the case of the latter, we 
may assign the same causes to like phenomena of the magnet. 

467. Mutual Action of Currents. 

1. If galvanic currents flow through parallel wires in the same 
direction, they attract each other ; if in opposite directions, they 
repel each other. These effects are shown in Fig. 250, where A B, 
A B r , turn toward each other, 
while C D, C D , turn away 
from each other. 

Hence, when a current flows 
through a loose and flexible 

FIG. 251 

helix, each turn of the coil at 
tracts the next, since the current ,_j ^ ^ 

moves in the same direction -^^i fe^ ^Jn 




through them all. In this way, a coil suspended above a cup of 
mercury, so as to just dip into the fluid, will vibrate up and down 
as long as a current is supplied. The weight of the helix causes 
its extremity to dip into the mercury below it ; this closes the 
circuit, the current flows through it, the spirals attract each other, 
and lift the end out of the mercury ; this breaks the circuit, and it 
falls again, and thus the movement is continued. 

2. If currents flow through two wires near each other, which 
are free to change their directions, the wires tend to become paral 
lel to each other, with the currents flowing in the same direction. 
Thus, two circular wires, free to revolve about vertical axes, when 
currents flow through them, place themselves by mutual attrac 
tions in parallel planes, as in Fig. 251, or in the same plane, as in 
Fig. 252. In the latter case, we must consider the parts of the 
two circuits which are nearest to each other as small portions of 
the dotted straight lines, c d and ef. 

FIG. 252. 

It appears, therefore, that galvanic currents, by mutual attrac 
tions and rejndsions, tend to place themselves parallel to each other 
in such a manner that the flow is in the same direction. 

Supposing the same to hold true of the molecular currents of 
magnets, this single law will satisfactorily account for the phe 
nomena of magnetic polarity. 

In the following articles these phenomena are considered in the 
order in which they are mentioned in Art. 465. 

468. Relations of Currents and Magnets to Each Other 
(1. Art. 465). It should be constantly borne in mind that when 
the north pole of a magnet turns toward a person, its currents cir 
culate from his right over to his left. 

1. When two solenoids, suspended as in Fig. 247, or when a 
solenoid and a magnet, or two magnets, are brought near each 
other, poles of different names attract, and those of the same name 



repel. For, when the magnets suspended from A and B (Fig. 253) 
are in the same line, it is seen that the currents are parallel and 

FIG. 253. 

flow in the same direction in all the corresponding parts ; and in 
Fig. 254, where they hang side by side, the nearer parts of the 

FIG. 254. 

FIG. 255. 

FIG. 256. 

currents are parallel and flow in the same direction. While in 
Fig. 255, where like poles are contiguous, the corresponding parts 
of the currents flow in opposite directions. 

2. When a magnet is suspended within a loop through which 
a current flows, if free to move 
it will place itself at right angles 
to the plane of the circuit, with 
the north pole pointing toward 
a person, when the current 
passes from his right over to 
his left (Fig. 256). Therefore, 
if the circuit is in a horizontal 
plane, the magnet turns its north 
pole downward, if the current 
flows as in Fig. 257, or upward 
if the current is reversed. 



3. When a magnet is brought near a closed circuit wire, 
as H (Fig. 258), it will place itself tangentially to a circle, x y z, 

FIG. 257. 

whose centre is in the wire, and its plane perpendicular to it. 
The part of the wire nearest to the magnet may be considered as 
a small portion of a loop around it, as in Fig. 25G. This tangen 
tial relation is maintained on all sides of the circuit, it being 
everywhere true that when the north pole is directed to a person, 
the current descends on the left, as if it had passed from the right 
over to the left. 

Comparing Figs. 257 and 258, it is evident that the current 
and the magnet may change places without disturbing their rela 
tive directions, it being understood that the current jloivs in the 
same direction in which the north pole points. 

469. The Galvanometer. Advantage is taken of the direc 
tive influence of a current on a magnet in the construction of the 
galvanometer (Fig. 259). When the 
coil consists of many convolutions of 
wire, a very feeble current passing 
through will deflect the needle from its 
north and south direction, and the 
amount of deflection serves as a measure 
of the galvanic force. Hence the name 
of the instrument. To render it still 
more sensitive, a second smaller needle, 
with poles reversed, attached to the same 
vertical wire, makes the first nearly 

astatic with relation to the earth. In making such a coil, the 
wire must be carefully insulated. This is generally done by wind 
ing it with silk thread. In the figure, the galvanometer is repre- 

FIG. 259. 



sen ted as covered by a bell-glass. The coil is seen beneath the 
graduated circle ; the deflected needle projects as a white line from 
within the coil, and directly above it is the needle, which nearly 
neutralizes the earth s influence upon it. 

470. Polarity with Respect to the Earth (2. Art. 465). 

It is believed that currents of electricity are constantly traversing 
the earth s crust, passing around it from east to west, and making 
the earth itself a magnet, with boreal magnetism developed at the 
north pole, and austral at the south pole. Thus the earth may be 
taken as the standard magnet, and both it and the currents around 
it control the polarity of the needle. For, as in Fig. 260, in order 

FIG. 260. 

that the current of the magnet may be parallel with the adjacent 
terrestrial current, and in the same direction with it, since the 
latter passes from east to west, the lower side of the former must 
also pass from east to west. But in order that this may be the 
case, the north pole of the magnet must point northward, and this 
it does when free to obey the directive influence of the earth. 

At first view, the earth currents from east to west seem to be 
in the wrong direction ; for that is from left over to right, to a 
person to whom the north pole points. This, however, is ex 
plained by recollecting that the magnetism of the north pole of 
the earth is the same as that of the south pole of a magnet (Art. 
380). For convenience, that end of a needle which points north 
is called the north pole ; but by the law of attraction between op 
posite poles, it must be unlike the north pole of the earth. There 
fore, the rule for the direction of currents around a magnet must 
be reversed when applied to the earth. 


The existence of currents traversing the earth s crust has 1 :i 
variously accounted for. The strong analogy between them in id 
those of thermo-electricity points to the heat of the sun as at least 
a very probable cause. 

471. Thermo-Electricity. Let a number of bars of bis 
muth (b) and antimony (a) be soldered together as in Fig. 2G1. 
!Now if the flame of a candle be carried 

around so as to warm the outer joints, a 
current of electricity will pass through 
the circuit from left to right, and will in 
fluence a needle near it just as any other 
current would do, flowing in the same di 
rection. Furthermore, it is only while the 
metals are unequally heated that the cur 
rent flows. "We may therefore suppose that 
the terrestrial current may be caused, in 
part at least, by the unequal heating of 

the heterogeneous substances composing the earth s crust, as the 
sun s heat is alternately poured upon and withdrawn from them 
once in every diurnal revolution. 

472. Magnetic Induction by Currents (3. Art. 465). 

Ampere accounted for the phenomena of magnetic induction by 
supposing that galvanic currents circulate through the molecules 
of all bodies, but in different directions, so that they mutually 
neutralize each other. That in a few substances, such as steel and 
iron, it is possible to control these currents and cause them all to 
flow in the same direction ; and that when this is done, the phe 
nomena of polarity ensue. 

Supposing this to be the correct explanation, tho effect of a 
galvanic current (and in fact of any method of magnetizing) is 
simply, by repulsion and attraction, to produce uniformity of 
direction among these magnetic currents. 

473. The Permanent and Temporary Magnet. When a 
current of sufficient strength is passed around a bar of well-tem 
pered steel, a permanent magnet of considerable power may be 

With soft iron, the result is a temporary magnet, which retains 
its magnetic properties only while the current is in motion. In 
either case the poles are always in the position which those of a 
needle would voluntarily assume if placed in the same relation to 
the current. 

474. Tho U-Magnet. Let a piece of soft iron, in the form 
of a horseshoe or the letter U (Fig. 262), be wound with a coil of 



FIG. 262. 

insulated copper wire whose extremities, W and w, are dipped in 
cups of mercury, in which are also dipped the electrodes + and 
of a battery. When all the wires are 
in metallic communication, the cir 
cuit is closed, and the current pass 
ing around the iron makes it a mag 
net ; and since to a person looking 
along the length of the helix the 
current passes from right over to 
left, the north pole is at JVJ and the 
south pole at S. As soon as the 
circuit is broken by lifting out of 
the mercury any one of the wires, 
the weight which was previously sus 
tained will fall, showing that the 
iron is no longer a magnet. 

FIG. 263. 

475. Helices. The form of coil or helix generally employed 
is shown in Fig. 263. Many hun 
dreds or even thousands of feet of 
insulated wire are wound around two 
bobbins, and through the centre of 
each passes a branch of the U-shaped 
iron; or, more frequently the central 
cores of iron are separate pieces, 
joined by a third one across two of 
the ends, and thus a U-magnet of 
modified form is obtained. By em 
ploying a fine wire coiled many times 
around the bobbins, a magnet of very 
great power may be formed, consider 
ing the weakness of the battery which 
furnishes the current. A magnet 
formed by the use of a small Bunsen 
cell has been known to lift five hun 
dred pounds, and with twenty Grove cells can be made to sustain 
a weight of three tons. 






476. Experiment Place a copper wire, a b (Fig. 264), near 
p n, the circuit wire of a battery. The current flowing through 
p n acts inductively on a b, decomposing its natural electricity. 
According to the general law of induction, the positive electricity of 

FIG. 264. 

a b is attracted in the direction of n, and the negative in the direc 
tion of p. In this experiment let the following facts be noticed : 

(1) The decomposition of the natural electricity of a 1 occurs 
at the instant of closing the circuit p n. 

(2) While the circuit remains closed, the current passing 
through it does not induce a current through a b. There is prob 
ably, however, an accumulation of positive electricity in the direc 
tion of b, and of negative in the direction of a ; for 

(3) When the circuit is broken, the natural electrical equilib 
rium of a b is instantly restored, and no further signs of a current 
can be detected until the circuit is again closed. 

The circuit is conveniently closed by dipping the end of the 
wire, e or d, in the cup of mercury m, and broken by removing it 
from the cup. 

477. The Induced and Inducing Currents. The sudden 
decomposition of the natural electricity mentioned in (1) of the 
preceding Article involves a momentary flow of the two electrici 
ties in opposite directions ; that is, a current is made to traverse 
the wire from a to b, that being the direction of the flow of the 
positive fluid (Art. 457). And the restoration of equilibrium 
mentioned in (3) involves a reversal of this flow ; that is, it pro 
duces a current which passes from b to a. 


These two currents in a b are called induced currents ; and the 
one in p n, to which they owe their origin, is called the inducing 
current. The presence, direction, and duration of the induced 
currents are indicated by the galvanometer g. 

The terms flow, current, accumulation of electricity, &c., when 
applied to the whole wire, have no significance except as they in 
dicate the resultants of those electrical disturbances which are 
probably confined to the individual molecules of the wire. 

478. Characteristics of Induced Currents. It is obvious 
that induced currents differ materially from the current of a bat 
tery which is uniform in direction and constant in intensity for 
an appreciable length of time. The following are the distinctive 
features of induced currents : 

(1) Induced currents are instantaneous. 

(2) They result from interruptions of the inducing current. 

(3) On closing the circuit, the direction of the resulting in 
duced current is opposite to that of the inducing current. 

(4) On breaking the circuit, the induced and inducing currents 
are in the same direction. 

479. Inducing and Induced Currents in one Wire. 

We have thus far considered only the inductive influence of the 
current on a wire exterior to its circuit. But the circuit-wire p n 
itself possesses its share of natural electricity, as well as a b. 

This is believed to exist independently of the galvanic current 
passing through it, and to be decomposed by that current. In 
order, therefore, to produce the preceding results with a single 
wire, let the circuit-wire be 
coiled as in Fig. 265. Each FIG. 265. 

spire is now acted upon induc 
tively by the galvanic current 
passing through the adjacent 
spires in the manner already 
described for separate wires. 
In addition to this mutual in 
ductive influence of the several 
spires on each other, it is prob 
able that the natural electricity of every portion of the wire is still 
further decomposed by the galvanic current passing through it 
For it is a noticeable fact that when a very long circuit-wire is 
employed, induced currents are obtained even though it be so 
nearly straight that no one portion can act inductively on another. 

The result of these several inductive actions is that when the 
circuit is closed and broken, regular induced currents are gen 
erated in it. And since these coexist for an instant of time with 


the inducing current, and pass through the same electrodes with it, 
it follows 

(1) That when the circuit is closed, the inducing current is 
partially neutralized, and has its intensity diminished by the in 
duced current which flows in a direction contrary to its own ; 

(2) That when it is opened, the induced current having now 
the same direction as the inducing current, reinforces it and aug 
ments its intensity. 

480. Mode of Naming Circuits and Currents. The phe 
nomena of induced currents were discovered by Faraday in 1832, 
and to him we owe the foregoing explanation of them. The fol 
lowing terms now in use were also introduced by him : 

The inducing current is called the primary current, and the 
wire it traverses the primary wire. Currents induced in the pri 
mary wire are called extra currents ; the one obtained on closing 
the circuit is the inverse extra current ; the one on opening it is 
the direct extra current (Art. 478, 3, 4). 

A wire exterior to the primary, as a b in Fig. 264, is a sec 
ondary wire, and the currents induced in it are secondary cur 

481. Currents Induced in Coils. Instead of straight 
wires or loose spirals, compact coils of carefully insulated wire are 
employed. Thus all parts of the wire are brought much nearer to 
each other, and the inductive influence is far more energetic. In 
deed, without a coil, the presence of induced currents can gener 
ally be detected only with a delicate galvanometer. The following 
experiments show the effects of coils : 

(1) Around a hollow wooden bobbin, Z (Fig. 266), coil about 
100 feet of No. 16 insulated copper wire. Let this be made a part 
of the circuit of a battery, as shown in the 
figure. This circuit is of course closed ^ FlG - 
when m and n touch each other. Now if 
m and n be held one in each hand and 
then separated, the body of the operator 
becomes a part of the circuit, and the pri 
mary current, not having sufficient inten- 
sity to pass through it, ceases. But the 
direct extra current passes through, producing a shock. When 
the wires are brought together again, the primary and inverse 
extra currents pass through the metallic circuit, and no shock is 

The more rapid the rate at which m and n are brought to 
gether and separated, the more decided are the results obtained. 



FIG. 267. 

To produce the most marked effect, attach a coarse file to one end, 
as m, and hold it in one hand while n is drawn rapidly over the 
ridges of its surface with the other. 

(2) Fig. 267 represents the same coil as Fig. 266, with the ad 
dition of a bundle of soft iron wires, w, inserted 

in the hollow bobbin. 

When the circuit is closed, these wires are 
magnetized that is, the Amperean currents sup 
posed to reside in them are made to circulate in 
the same direction as the battery current (Art. 

And since the appearance and disappearance 
of these magnetic currents are simultaneous with 
the appearance and disappearance of the primary 
current, they augment the effects of the latter, and the resulting 
extra currents are of greater intensity. 

The effect of soft iron in the primary coil is an observed fact, 
and the above is the way in which Faraday accounted for it on 
the basis of Ampere s theory. 

(3) Let the primary coil and bundle of wires of the preceding 
figure be placed within a secondary coil, d (Fig. 268), from which 
it is carefully insulated. This 

secondary coil should be made 
of wire much greater in 
length and smaller in diam 
eter than that of which the 
primary coil is made. For 
instance, let it consist of 1500 
feet of No. 35 insulated cop 
per wire. When the ends of 
this wire, 7i, h 9 are held one 
in each hand, every time the 
primary circuit is interrupted, 
a secondary current traverses 
the secondary circuit of which the person forms a part. The re 
sulting shocks will be quite appreciable, though the primary cur 
rent be produced by only a single small cell. 

As in the first experiment, the effect on the person will become 
more marked as the interruptions increase in frequency. 

In the third experiment, the magnetic currents of the iron core 
add their inductive influence, as already explained, to that of the 
primary current, thus increasing the intensity of the secondary 

The effect of the extra currents also should not be overlooked. 
As these traverse the primary coil, alternating with each other in 

FIG. 268. 


FIG. 269. 

direction, they materially modify the effects of the primary and 
magnetic currents, which are uniform in direction. 

482. Ruhmkorff s Coil. The celebrated Ruhmkorff coil 
(Fig. 269) is not essentially different from the one just described, 
except in having (1) an ar 
rangement for producing a 
continued and rapid succes 
sion of interruptions in the 
primary current, (2) a com 
mutator, or key, by which 
when desired the primary 
current may be stopped or 
its direction reversed, and 
(3) a condenser to neutralize 
the effects of the extra cur 
rents. The condenser con 
sists of sheets of tin-foil in 
sulated by oiled silk. They 
are placed out of sight, in 
the base of the apparatus, 
and are so connected w r ith 

the primary wire that the extra currents pass into them. Owing 
to this diversion of these interfering currents the efficiency of the 
coil is very much increased. 

483. Power of the Ruhmkorff Coil. The efficiency of a 
Ruhmkorff coil depends largely on complete insulation; and, in 
different coils, varies greatly with the length and fineness of the 
secondary wire. 

To secure insulation, the wires are (as usual) wound with silk 
thread, then each individual coil around the axis, is separated from 
the succeeding one by a layer of melted shellac, and lastly a cylin 
der of glass is placed between the primary and secondary -coils. 

With regard to the secondary coil, one of the largest size some 
times contains sixty miles of the finest copper wire. "With such 
an apparatus, though the primary current be produced by only 
two or three Bunsen cells, the secondary currents are of such in 
tensity that sparks eighteen inches in length, and of great bril 
liancy, may be obtained ; and indeed, all the tension effects of a 
large electrical machine, as well as the quantity effects of a power 
ful galvanic battery, may be reproduced. 

Great care should be taken in handling an induction coil of 
this size, for the shock resulting from its discharge through the 
body would be dangerous, and might possibly prove fatal. 



FIG. 270. 

484. One Coil moved into, and out of, another. In all 

that has preceded, the interruptions of the primary current have 
been supposed to take place instantaneously. If these interrup 
tions are gradual, the resulting induced currents remain the same 
in direction as before, but vary in intensity and duration. 

Thus, if the primary coil c (Fig. 270) be made to fit loosely in 
the secondary coil d, and then be moved up and down (the pri 
mary circuit remaining 
closed), it will be found 

(1.) That each inser 
tion and removal of it cor 
responds, the one to a 
gradual closing, the other 
to a gradual opening of the 
primary circuit the result 
of the former being an in 
verse secondary current, of 
the latter a direct seconda 
ry current. 

And since a continuous 
motion of the primary coil 
produces a continuous se 
ries of instantaneous sec 
ondary currents with no 
appreciable interval between them, it will be found 

(2.) That the secondary currents are continuous in effect as 
long as the motion of the primary coil is continuous; 

(3.) That their intensity varies with the rate of motion of the 
primary coil, diminishing or increasing as that is moved slowly or 
rapidly ; from which it follows 

(4.) That they cease whenever the primary coil is brought to a 
state of rest in any position. 

485. Changes of Intensity in the Primary Current All 

the results just mentioned may be obtained if, instead of changing 
the position of the primary coil, as above, it remain at rest while 
a corresponding series of variations be produced in the primary 
current, an increase of intensity in that corresponding to an 
insertion of the coil, and a decrease to a removal of it. 


486. Magneto-electricity. Faraday reasoned that if cur 
rents could induce magnetism, a magnet ought to induce currents. 



FIG. 271. 

This he found to be the case, and thus discovered a new branch 
of physical science, to which he gave the name of magneto-elec 

If a magnet be used instead of the primary coil in Fig. 270, all 
the phenomena mentioned in Art 484 may be reproduced. 

Thus, with the coil and magnet in 
Fig. 271, we obtain the following re 
sults : 

(1.) "When the magnet is alternately 
inserted in and withdrawn from the coil, 
the latter is traversed by induced cur 
rents alternating with each other in di 

(2.) These currents are continuous 
while the magnet is in motion. 

(3.) Their intensity diminishes or 
increases as the magnet moves slowly or 

(4.) They cease when the motion of the magnet ceases. 

487. Explanation of the foregoing Phenomena. On the 

basis of Ampere s theory, the correspondence of these phenomena 
with those of Art. 484 can readily be accounted for. For the mag 
net may be considered a true primary coil, its magnetic currents 
corresponding to the primary current in Fig. 270. With regard 
to them, the induced currents are regular inverse and direct sec 
ondaries ; for in any given case they will be found to have the 
same direction as those induced by a primary current whose di 
rection corresponds with the supposed direction of the magnetic 

It will be seen at once what a strong argument is here fur 
nished in favor of Ampere s 

* 488. An Iron Core, 
changing its Magnetic 
Intensity. Replace the 
magnet (Fig. 271) by a bar 
of soft iron inserted in the 
coil, and let a magnet be 
alternately brought near 
this, and removed from it, 
as in Fig. 272. The same 
results will be obtained as 
in the preceding series of 
experiments. The proxim- 

FIQ. 272. 



ity of the magnet induces magnetism in the soft iron, and its 
motions to and fro produce variations in this induced magnetism 
corresponding precisely with the varying intensity of the primary 
current mentioned in Art. 485, and, as might be expected, the 
results are the same. 

In this experiment, it is obviously immaterial whether the coil 
be at rest and the magnet be moved, or the magnet be at rest and 
the coil be moved. The latter method is adopted in some mag 
neto-electric machines. 

489. Clarke s Magneto-electric Machine. In front of 
the poles of the U-magnet, A (Fig. 273), is revolved the armature, 

FIG. 273. 

consisting of the two "bobbins, B, B ! , which are coils of fine wire 
with cores of soft iron. These cores are joined to each other and 
to the axis of rotation by the bar of soft iron, V, and motion is 
communicated by the multiplying wheel and band at W. 

As One of the bobbins passes before a north pole while the 
other is passing before a south pole, the resulting induced cur 
rents are relatively of contrary directions ; but as one of the coils 
is always right-handed and the other left-handed, the currents 
passing through them at any given instant have the same abso 
lute direction, so that the two coils act as one. 



FIG. 274 

Fig. 274 shows the poles of the fixed magnet, and the direction 
in which the armature revolves. The maximum magnetization 
of the soft iron cores occurs when the bobbins are directly in front 
of JV^and & While they move through 
the first and third quadrants they are 
losing their magnetism, and while mov 
ing through the second and fourth they 
are acquiring that of the contrary kind. 
The resulting induced currents will thus 
be direct and inverse to contrary kinds 
of magnetism, and will therefore have 
the same absolute direction. But as the 
bobbins pass from the second quadrant 
to the third, and from the fourth to the 
first, they lose the magnetism just ac 
quired, and the induced currents change 
from inverse to direct with reference to 
the same kind of magnetism, and there 
fore become reversed in absolute direc 

Hence the semi-revolutions of the 
armature on opposite sides of a line 

joining the poles of the permanent magnet produce currents of 
contrary directions. 

490. The Commutator. Fig. 275 is an enlarged view of the 
outer end of the axis (Fig. 273), and shows the commutator, or 

FIG. 275. 

arrangement by means of which the contrary currents just men 
tioned are made to furnish one or rather a series of currents 
flowing in the same direction. 

Two pieces of brass, m and n, are insulated from each 
other by being fastened to an ivory ring, i, around the axis. They 
are so connected with the wires of the coils that at any given in 
stant both coils present to m one kind of polarity, and to n the 


other ; that is, they are made the poles of the two coils acting as 
one. Against m and n press two springs, 1) and c, which are 
brought into communication with each other through the plates 
P and P , and the handles h and h , whenever the latter are joined. 
When b and c press against m and n respectively, it will he seen 
that the circuit is complete. Let us suppose that the induced cur 
rent is passing through it in the direction indicated by the arrows. 
When the armature has revolved through 180 from its present 
position, m and n will have changed places but they will also have 
changed polarities (Art. 489). Therefore n presents to b the same 
polarity which m did, and hence there is no change in the direc 
tion of the current through I and c. 

491. Effects of Rapid Revolution. The intensity of the 
induced currents of this machine, as also the rapidity with which 
they succeed each other, is regulated by the rate of revolution of 
the armature. When this is rapidly revolved, they produce all the 
effects of a single voltaic current, so that the apparatus may be 
used as a galvanic battery with h and h r for its electrodes. At the 
same time its physiological effects are most remarkable, the shocks 
becoming unendurable when it is revolved with great rapidity. 
The shocks are more powerful when a third spring, a (Fig. 274), 
is attached to the plate P , near c. 

492. Large Machines. In large magneto-electric machines 
of this kind, increased efficiency is obtained in two ways : 

First, by multiplication of magnets. In Nollet s machine, con 
structed in 1850, 192 magnetized steel plates are so combined as 
to make 40 powerful U-magnets. These are arranged in eight 
rows around the circumference of a large iron frame inside of 
which revolve sixty-four bobbins. 

Second, by multiplication of currents. In Wild s machine, con 
structed on this plan, the induced currents first obtained, instead 
of being directly utilized, are passed through the coils of a large 
electro-magnet. Before the poles of this a second armature re 
volves, and the resulting induced currents are far more powerful 
than the first. These may in turn be made to magnetize a second 
electro-magnet before the poles of which a third armature re 
volves, &c. 

Currents of very great intensity may be obtained from either 
of these machines. The motive power employed is generally a 
steam-engine of from one to fifteen-horse power. 




493. Classification. The applications of Galvanic electricity 
in the arts and sciences, as well as in the affairs of every day life, 
are eminently practical. They may be classified according to the 
way in which are utilized those molecular forces whose resultant 
is known as the current. 

It may be stated in general that these applications are made 
either within the circuit, or exterior to it. 

Within the circuit. Here the electrical force is employed (I) 
directly, or (II) by being first made to produce the effects light 
and heat, which are then applied as desired. 

Without the circuit. Here it is employed indirectly (III) by 
being made to reappear as mechanical motion through the inter 
vention of the kindred force, magnetism. 

Examples will be given of each. 


494. Electrolysis. When a current is passed through a bi 
nary compound (i e., one containing two elements), the compound 
is decomposed, one of its elements appearing at the positive elec 
trode, the other at the negative. 

For instance, water, consisting of the two gases oxygen and 
hydrogen, is thus decomposed. 

In the bottom of the dish D (Fig. 276), partly filled with water, 
are fastened p and n, the 
platinum electrodes of a bat 
tery. Over these are placed 
two tubes, and //, full of 
water. On closing the cir 
cuit, oxygen rises from p 
into 0, and hydrogen from 
n into H. 

Electrolysis is of the ut 
most importance in chemis 
try. Thus, the preceding 
experiment gives a correct 
analysis of water, and if oxygen had been previously unknown, 


would have been the means of its discovery. In this way were 
discovered several of the metallic elements. 

No less important are its applications in the arts. For when a 
solution of a metallic salt is subjected to the action of the current, 
it is decomposed and a permanent film of the metal is deposited 
on any suitable material placed so as to receive it. The process is 
then called electro-metallurgy, or electro-plating. 

495. Electro-plating. The bath (Fig. 277) contains a satu 
rated solution of blue vitriol (sulphate of copper). In this is sus 
pended by wires from the metallic rod D a plate of copper C, and 
from B (also metallic) the 

cast of a medal m, which FlG - 

is to be coated with copper. 
Connect D with the posi 
tive electrode of a battery 
and B with the negative. 
The current passing 
through the solution re 
moves from it particles of 
copper and deposits them 
on m. Those taken from 
the liquid are replaced by others taken from C, which is thus 
gradually wasted away, and the solution is kept saturated. 

If the bath contains a solution of gold, and C is replaced by 
a piece of gold and m by a silver cup, the cup will be electro-gilded. 
Electro-silvering is an analogous process. 

To* produce in any case a firm and even coating, the process 
must be allowed to proceed slowly by the employment of a weak 
current. On a small scale a single cell is sufficient. In large es 
tablishments a magneto-electrical machine turned by steam has 
been successfully and economically used. 

496. Electrotyping. By taking proper precautions, the cop 
per film deposited on m may be removed, and its surface will be 
found to present an exact fac-simile of the medal of which m is an 
impression. Therefore if m is an impression of the type from 
which a page is printed, when the copper has been removed and 
stiffened by melted lead (or some alloy) poured over its under sur 
face, it may be used in the printing-press instead of the type. It 
is then called an electro-type plate, and when not in use may be 
preserved indefinitely for succeeding editions, while the type of 
which it is a copy can be distributed and used for other purposes. 

497. Medicinal Applications. The shocks produced by 
the passage of interrupted currents through the system have al- 



ready been alluded to. In certain ailments these shocks, when 
properly applied, have been known to produce beneficial results. 
On the other hand, great injury has resulted from their misappli 
cation. Hence they should be employed as remedies only under 
the direction of a reliable physician. 

The familiar medical magneto-electrical machine, which comes 
compactly stored in a box ten inches long and about four inches 
square at the end, does not differ essentially from Clarke s (Art. 
4S9), except in lacking the commutator, so that its currents pass 
through the body alternating with each other in direction. 


498. Light by the Electric Current. The electric light 
may be advantageously employed for brilliant illumination on 
special occasions ; also where a strong penetrating light is needed, 
as in light-houses, or for signals between ships. For exhibiting 
to an audience magnified images of small objects (as with the pro 
jecting microscope) it has no superior ; and to the physical experi 
menter the various colors it assumes on passing through highly 
rarefied gases of different kinds are of great interest. But as yet 
it cannot compete with gas-light for ordinary illumination. 

To obtain the most brilliant effects carbon electrodes must be 
employed, and as these are constantly changing in length they 
must be kept at a uniform distance apart by machinery. The 
flame is not straight, but curved, as in Fig. 278, and is called the 
voltaic arc. To ob 
tain it, the electrodes ** m 
must first be made 
to touch each other. 
With 92 Bunsen ele 
ments the light has 
been found to pos 
sess more than one- 
third the intensity of 
direct sunlight. 

499. Heat by 
the Electric Cur 
rent. The heat as 
well as the light of 
the voltaic arc is in 
tense. In the labo 
ratory it is employed 
to deflagrate and vol- 



atilize refractory substances. When the lower electrode is hol 
lowed out in the form of a cup, a piece of platinum (one of the 
least fusible of metals) placed in it is melted like wax in a candle, 
and a diamond, the hardest of known substances, is burnt to a 
black cinder. To produce either of these results, a battery of 
great power must be used. 

Metals may also be deflagrated by being made part of the cir 
cuit in the form of very fine wires. They are thus employed to 
spring mines in time of war, or in blasting rocks. In Fig. 279, B 
is a box full of fulminating powder, and w is a very fine platinum 

FIG. 279. 

wire, about | in. in length, fastened to p and n, which are insu 
lated copper wires extending to a battery situated at any conve 
nient distance. B and its contents are the fuse which is inserted 
in the powder to be fired. When a moderately strong current 
passes through w, it is heated sufficiently to ignite the fuse, and 
the powder explodes. 


500. Made through the Medium of Induced Magnetism. 

As has been seen in preceding experiments, magnetism may be 
induced in a piece of steel or iron by the current passing through 
a circuit which is near it. Hence induced magnetism and its ap 
plications are results obtained outside of the circuit. 

This magnetism may be utilized directly. For compass and 
galvanometer needles are ordinarily made by placing a steel needle 
in a helix through which a current is sent (Art. 473). 

But its most numerous and important applications are in the 
way of mechanical movements. All these are modifications of the 
simple rising and falling of the armature of a U-magnet, men 
tioned in Art. 474. Thus, the 
armature A (Fig. 280) is lim 
ited in its fall by the metallic 
base B, so that it is within the 
influence of M the next time 
that it becomes a magnet. 
Hence, when the circuit is 
closed and broken at n p, the 

FIG. 280. 


end A of the lever L rises and falls. It is evident that the cor 
responding motions of the end E may be applied in a variety ut 
ways. A few of these are described in the following articles. 

501. Electro-magnetic Engine.^ 7 may be attached to n 
vertical arm, and that to the crank of a fly-wheel (Fig. 281), and 
the interruptions of the current may be made automatic by con 
necting p with L, and n 

with B. When A rests on FlG - 281 

B the circuit is closed ; M 
becomes a magnet, and A is 
attracted by it ; but as soon 
as A rises from B the cir 
cuit is opened, M no longer 
attracts it, and it falls back, 
only to close, the circuit 
again and repeat the same 
movements as before. The 
tendency of this is to pro 
duce a rotary motion in the 
fly-wheel, and the apparatus 
involves the principle of a 
single-acting engine. With a second electro-magnet, and a some 
what different arrangement of parts, an actual double-acting en 
gine may be constructed. 

Various forms of this engine have been constructed and exhib 
ited as curiosities, or used where expense was not regarded. But 
it cannot compete with the steam-engine as long as zinc and acids 
cost so much more than coal. 

502. Electro-magnetic Telegraph, To our countryman, 
Prof. S. F. B. Morse, is due the credit of the erection of the first 
telegraph line in the United States. It extended from Baltimore 
to Washington, and went into operation in 1844. 

Communication in various ways by means of electricity be 
tween places a few miles apart was not unknown in Europe before 
that time, and several ingenious systems have appeared since. 
One of these is Wheatstone s, which is commonly used in England. 
But the Morse system has been very generally preferred on ac 
count of its greater simplicity and efficiency, and it is now widely 
used in the United States and on the continent of Europe, where 
it is known as the American system. The principle of its opera 
tion is as follows : 

Let E of Fig. 280 be furnished with a style e (Fig. 282) directly 
over which is the groove on the surface of a solid brass roller c. 
Between c and e is the long paper ribbon R R. Also let A be 



placed above J/and be furnished with a spring s to raise it as far 
as the screw i allows when it is not attracted by M. When the 
circuit is closed, A is attracted and e rises and forces the paper 

FIG. 282. 

into the groove, producing a slight elevation on its upper surface. 
The ribbon is pulled along at a uniform rate in the direction of 
the arrow by clockwork (not shown in the figure), so that when 
the circuit remains closed for a little time, a dash is marked on 
the paper by e ; when it is closed and instantly opened, the result 
is a dot or rather a very short dash. Spaces are left between 
these whenever the circuit is opened. Combinations of these dots, 
dashes, and spaces, all carefully regulated in length, compose the 
letters of the alphabet. Spaces are also left between the letters, 
and longer ones between words. 

By lengthening the circuit wire, it is evident that the person 
who sends the message at n p, and the one who receives it at E, 
may be miles apart, and the transmission will be almost instanta 
neous owing to the rapid passage of the current. 

The essential parts of this system, or indeed of any system, are 
a communicator at n p, an indicator at E, and a wire extending 
from one to the other. 

503. The Connecting Wire. It was at first supposed that 
a complete metallic circuit was necessary, hence a return wire was 

FIG. 283. 

employed. But this was rejected when it was found that the 
earth could be used as a part of the circuit, as shown in Fig. 283. 


/S and $ are the terminal stations, and s is one of the way stations 
which may occur anywhere along the line. At every station both 
a communicator, ( , and indicator, 7, are introduced into the cir 
cuit, so that messages can be both sent and received. 

504. The Communicator. This consists of a lever, I (Fig. 
284), and anvil, a, both of brass, and insulated from each other. 

FIG. 284. 

The anvil connects with the line wire W, and I with the rest of the 
circuit through IT , and W W of the next figure. (See also Fig. 
283.) The end of I is depressed by the finger of the operator on the 
insulating button b, and is raised by the spring s when the pressure 
is removed. The former movement closes the circuit, the latter 
opens it, and by a succession of these the message is sent. 

When the communicator is not in use, the brass bar k hinged 
to the base of I is pressed into contact with a. This closes the 
circuit for other stations on the line, and hence k is called the cir 
cuit closer. The whole apparatus is called the key. 

505. The Indicator. This consists of two parts, (1st) the 
relay, and (2d) either the register, or the sounder. 

The first part, called the relay, or the relay magnet (Fig. 285), 
consists of an electro-magnet, an armature, a lever, and a spring, 
the same as in Fig. 282, except that the electro-magnet is horizon 
tal, and the other parts correspond in position. The tension of 
the spring 5 is regulated by the screw and milled head h, and J/ 
is adjusted by a similar screw (between W and W" in the figure), 
which slides it along the grooved way X. One end of the coil 
wire passes out through W to W of Fig. 284. The other end 
connects at II " with one pole of the battery if it is at S (Fig. 283), 
with the earth if it is at /S", or with the line wire to the next sta 
tion if it is at s. 

The reason for introducing the relay is this: The current 
from the preceding station has become too feeble to cause indenta 
tion of the paper by the style, and thus make a visible record, or 
even to produce a distinct sound of the armature upon the mag 
net for reading messages by the ear. The relay is therefore con 
trived for employing this feeble current to close and open the 



circuit of a local battery, whose current is powerful enough to dc- 
liyer messages in either form, or even in both forms at once. All 
which the weak current of the distant buttery has to do is to cause 
the armature A to move toward the magnet M till the top of L 
touches the screw N, and thus closes the circuit of the local bat 
tery. When the current ceases, a delicate spring, s, draws L "back 
from contact with N, and breaks the circuit of the local batterv. 

By the adjustments above described, the distance through which 
L moves, and the force of the spring s, may be made as small as 
the operator pleases. 

506. The Register and the Sounder. The second part of 
the indicator is either a register, or a sounder, according as mes 
sages are to be addressed to the eye or to the ear. The register 
(Fig. 282) has been already described ; the current of the local 
battery close at hand has force enough to cause visible indenta- 

FIG. 286. 
o f 

tions in the paper whenever the lever is drawn to the magnet ; 
and this record can be read at any time subsequently. Within a 
*few years a modified form of the register has come into use, and 
is called the sounder. In this the end of the lever L (Fig. 286), 


instead of being furnished with a style, is made to strike against 
the two screws, JV , . The downward dick is a little louder 
than the upward one, and so the beginning and end of each dot 
or dash are distinguished from each other. Many operators learn 
from the first to read by the car, and have never used a register. 

Whether a register or a sounder is employed, its coil wire is 
entirely distinct from the line wire, and belongs only to the local 
battery. The circuit of this battery may be traced (Figs. 285, 286) 

from the positive pole through I, L, JV", n, 0, 0, ri, the coil of 

the sounder, and n", to the negative pole. The binding screws, Z, 
n, n 9 n", are connected with their respective levers, or contact 
screws, by insulated wires concealed in the bases. When A is at 
tracted by J/, L touches N 9 and the circuit is closed ; when it is 
withdrawn by s, the circuit is opened, because is insulated. 
Hence the motions of L and L are simultaneous. 

Since the relay is always in the main circuit, it communicates, 
by means of the local current, to the operator at whose station it 
is, all the messages sent between any two stations on the line, in 
cluding those which he himself sends. Hence, if his own indi 
cator does not operate while he is at work, he knows that his mes 
sage is not passing over the line, owing to some break in the 

507. Repeaters. On a well insulated wire the weakness of 
the current at the distance of a few miles from the battery is 
mainly due to the resistance of the wire. The nature of this re 
sistance is unknown, but it is subject to the same law as ihe fric 
tion of a fluid along the interior of a tube ; it varies directly as 
the length^ and inversely as the diameter. Hence telegraph wires 
of considerable thickness are employed, and even then, after a cer 
tain number of miles, varying according to strength of battery, 
insulation, etc., the current will not work even a relay. Before 
reaching that point, therefore, the wire is allowed to pass into the 
ground, and so complete the circuit. 

To pass a message beyond the place at which the current will 
only work a relay, N and L 1 are made parts of a new circuit 
called a repeater, which is closed and opened simultaneously with 
the preceding one by the motions of L , just as a local circuit is 
worked by L (Art. 505). 

On the 28th of February, 18G8, signals were sent through 
from Cambridge, Mass., to San Francisco, by the employment of 
thirteen repeaters. The time occupied by the signals in going 
and returning (making about 7,000 miles) was three-tenths of a 
second allowance being made for the coil wires of the electro 
magnets through which the current passed. 


508. Atlantic Telegraph Cable. This cable stretches a 
distance of 3,500 miles, and from the nature of the case is a con 
tinuous wire, so that it cannot be advantageously worked by the 
Morse apparatus. The indicator employed is a sensitive galva 
nometer needle which is made to oscillate on opposite sides of the 
zero point by the passage through it of currents in opposite direc 
tions. But to reverse the direction of the current throughout the 
whole length of the cable is a slow process. For the cable is an 
immense Ley den jar, the surface of the copper wire (amounting to 
425,000 sq. feet) answering to the inner coating, the water of the 
ocean to the outer, and the gutta-percha between the two to the 
glass of an ordinary jar. A current passing into it is therefore 
detained by electricity of the contrary kind induced in the water, 
and no effect will be produced at the further end until it is 

This very circumstance, at first considered a misfortune, is now 
taken advantage of in a very simple and ingenious manner to 
facilitate the transmission of signals. The current is allowed to 
pass into the cable till it is charged then, ivithout breaking the 
circuit, by depressing a key for an instant, a connection is made 
between it and a wire running out into the sea; that is, between 
the inner and outer coatings. This partially discharges it, and 
the needle at the other end is deflected. "When the key is raised 
the discharge ceases, the current flows on as before, and the needle 
is deflected in the opposite direction. 

It is said that after this plan was adopted, twenty words could 
be sent through the cable per minute, whereas only four per min 
ute could be sent before. The greatest speed thus far attained on 
land wires is believed to have been the transmission in one in 
stance of 1,352 words in thirty minutes between New York and 
Philadelphia in 1868. 

509. Fire-Alarm Telegraph. Recurring again to the stand 
ard (Fig. 280), the end E may be so connected with machinery as 
to cause the striking of a bell in a distant tower whenever the cir 
cuit is closed at n p. In our large cities boxes are placed at con 
venient points, each containing a crank, or lever, by which the 
circuit may be closed and the fire-bell rung. Thus, by previously 
arranged signals, the locality of the fire is immediately made 
known at the various engine-houses. 

510. Chronograph. This is used in observatories for record 
ing the passage of stars across the meridian. Imagine the circuit 
of Fig. 282 to be closed and instantly broken again, by a clock 
pendulum at the end of every second. As the paper, R R, moves 
uniformly, dots are made on it at equal distances from each other, 



each of which distances, therefore, represents one second. The 
observer has a key, by which also he closes the circuit for an in 
stant when a certain star passes the meridian. The dot thus made 
shows, by its situation between the two nearest second dots, at 
what fraction of the second the transit occurred. 

In practice, however, the record is more conveniently made on 
a large sheet of paper, which is wrapped tightly around a cylinder. 
The clock-work, which revolves the cylinder, also moves the re 
cording pen in a line parallel to its axis. By these two motions, a 
spiral ink-line is traced on the paper. At the end of every beat 
of the observatory clock, the closing of the circuit gives the pen a 
momentary lateral movement, by which a slight notch is made in 
the line, A similar notch is made by the touch of the key, when 
the observer perceives the star on the meridian wire of the tele 
scope. Fig. 287 represents a portion of the sheet after its removal 

FIG. 287. 

from the cylinder ; , #, c, d, &c., are the second marks ; x, y, z, 
&c., are transit records. The ratio m y : m n shows what fraction 
of the second m n has elapsed when the transit y occurs. 


E A. T . 



511. Nature of Heat. Heat is another of those agencies 
which have been regarded as imponderable substances. It was 
said to emanate in straight lines from the sun and from bodies in 
combustion, to cause the sensation of warmth when it strikes us, 
to expand bodies when it enters them, to raise their temperature, 
&c. But there is abundant reason for believing that heat consists 
of exceedingly minute and rapid vibrations of ordinary matter and 
of the ether which fills all space. It is to be regarded as one of 
the modes of motion, which may be caused by any kind of force, 
and which may be made a measure of that force. Heat affects 
only one of our senses, that of feeling. Its increase produces the 
sensation of warmth, and its diminution that of cold. 

512. Expansion and Contraction by Heat and Cold. 

It is found to be a fact almost without exception, that as bodies 
are heated they are expanded, and that they contract as they are 
cooled. It is easy to conceive that the vibratory motion of the 
several molecules of a body compels them to recede from each 
other, and to recede the more as the vibration becomes more vio 
lent. Although the change in magnitude is generally very small, 
yet it is rendered visible by special contrivances, and* is made the 
means of measuring temperature. 

513. The Thermometer. This instrument measures the 
degree of heat, or the temperature, of the medium around it, by 
the expansion and contraction of some substance. The substance 
commonly employed is mercury. The liquid, being inclosed in a 
glass bulb, can expand only by rising in the fine bore of the stem, 
where very small changes of volume are rendered visible. A 


scale is attached to the stem for reading the degrees of tempera 

The graduation of the thermometer must begin with the fixing 
of two important points hy natural phenomena, the freezing and 
boiling of water. When the bulb is plunged into powdered ice, 
the point at which the column settles is the freezing point of the 
thermometer. And if it is placed in pure boiling water under a 
given atmospheric pressure, the mercury indicates the boiling point. 
Between these two points, namely 32 and 212 F., there must be 
180, and the scale is graduated accordingly. As the bore of the 
tube is not likely to be exactly equal in all parts, the length of the 
degrees should vary inversely as the area of the cross-section. 
This is accomplished by moving a short column of mercury along 
the different parts and comparing the lengths occupied by it. The 
degrees in the several parts must vary in the same ratio. 

514. Different Systems of Graduation. There are in use 
three kinds of thermometer scale, Fahrenheit s, Reaumur s, and 
the Centigrade. In Fahrenheit s, the freezing point of water is 
called 32, and the boiling point 212 ; in Reaumur s, the freezing 
point is called 0, and the boiling point 80 ; in the Centigrade, 
the freezing point 0, and the boiling point 100. In a scientific 
point of view, the Centigrade is preferable to either of the others, 
but Fahrenheit s is generally used in this country. The letter F., 
E., or C., appended to a number of degrees, indicates the scale 
intended. In this country, F. is understood if no letter is used. 

515. To Reduce from one Scale to Another. Since 
the zero of Fahrenheit s scale is 32 below the freezing point, while 
in both of the others it is at the freezing point, 32 must always 
be subtracted from any temperature according to Fahrenheit, in 
order to find its relation to the zero of the other scales. Then> 
since 212 -32 ( = 180) F. are equal to 80 R.,and to 100 C., the 
formula for changing F. to R. is ^ (F. 32)=R. ; and for changing 
F. to C., it is (F.-32) = C. Hence, to change R, to F., we have 
-J R. + 32^F.; and to change C. to F., J C. + 32 = F. 

Mercury congeals at about 39 F. ; therefore, for tempera 
tures lower than that, alcohol is used, which does not congeal at 
any known temperature. 

516. Expansion of Solids. When the expansion of a solid 
is considered simply in one dimension, it is called linear expan 
sion; in two dimensions only, superficial expansion; in all three 
dimensions, cubical expansion. 

The linear expansion of a metallic rod is readily made visible 
by an instrument called the pyrometer, which magnifies the mo- 

312 HEAT. 

tion. The end A of the rod A B (Fig. 288) is held in place by a 
screw. The end B rests against the short arm of the lever (7, the 
longer arm of which bears on the arm D of the long bent lever 

FIG. 288. 


D E\ this serves as an index to the graduated arc E F. The long 
metallic dish G G, being raised on the hinges HE, so as to en 
close the bar A B, and then rilled with hot water, the bar instantly 
expands, and raises the index along the arc E F. 

517. Coefficient of Expansion. The coefficient of linear 
expansion of a given substance is the fractional increase of its 
length, when its temperature is raised one degree. But since this 
increase is generally somewhat greater at higher temperatures, the 
coefficients of expansion given in tables usually refer to a temper 
ature at or near the freezing point of water. Thus, the coefficient 
of expansion for silver is 0.00001061 ; by which is meant that a 
silver bar one foot long at 32 F. becomes 1.00001061 ft. in length 
at 33 F. 

The coefficient of superficial expansion is twice, and that of 
cubical expansion three times as great as the coefficient of linear 
expansion. For, suppose c to be the coefficient of linear expan 
sion ; then if the edge of a cube is 1, and the temperature is raised 
1, the edge becomes 1+c, and the area of one side becomes 
(l+c) 2 =l+2c + c 2 , and the volume (l+c) 3 =l+3c + 3c 2 -f c 3 . But 
as c is very small, the higher powers may be neglected, and the 
area is 1 + 2c, and the volume is 1 + 3c ; that is, the coefficient of 
superficial expansion is 2c, and that of cubical expansion is 3c, as 
stated above. 

518. The Coefficient of Expansion differs in different 
Substances. Copper expands nearly twice as fast as platinum; 
the ratio of expansion in steel and brass is about as 61 to 100. 
This ratio is employed in the construction of the compensation 
pendulum (Art. 171). The same is sometimes used also to render 
constant the length of the rod with which the base line of a trigo 
nometrical survey is measured. 


If two thin slips of metal of different expansibility be soldered 
together so as to make a slip of double thickness, it will bend one 
way and the other by changes of temperature. If it is straight 
at a certain temperature, heating will bend it so as to bring the 
most expansible metal on the convex side ; and cooling will bend 
it in the opposite direction ; and the degree of flexure will be ac 
cording to the degree of change in temperature. Compensation 
in clocks and watches is sometimes effected on this plan. If the 
compound slip has the form of a helix, with the most expansible 
metal on the inside, heating will begin to uncoil it, and cooling, 
to coil it closer. A very sensitive thermometer, known as 13 re- 
guet s thermometer, is constructed on this principle. 

519. The Strength of the Thermal Force. It is found 
that the force exerted by a body, when expanding by heat or con 
tracting by cold, is equal to the mechanical force necessary to ex 
pand or compress the body to the same degree. The force is there 
fore very great. If the rails were to be fitted tightly end to end 
on a railroad, they would be forced out of their places by expan 
sion in warm weather, and the track ruined. The tire of a car 
riage wheel is heated till it is too large, and then put upon the 
wheel; when cool, it draws together the several parts with great 
firmness. In repeated instances, the walls of a building, when 
they have begun to spread by the lateral pressure of an arched 
roof, have been drawn together by the force of contraction in cool 
ing. A series of iron rods being passed across the building through 
the upper part of the walls, and broad nuts being screwed upon 
the ends, the alternate bars are expanded by the heat of lamps, 
and the nuts tightened. Then, when they cool, they draw the 
walls toward each other. The remaining bars are then treated in 
the same manner, and the process is repeated till the walls are 
restored to their vertical position and secured. For a measure of 
the force of heat see Art. 555. 

520. Expansion of Liquids. It has already been noticed 
that mercury and alcohol expand by heat, and are therefore used 
in thermometers for measuring temperature. These are the best 
liquids for such a purpose, because their temperature of congela 
tion is very low. 

As liquids have no permanent form, the coefficient of expan 
sion for them is always understood to be that of cubical expansion. 
There is a practical difficulty in the way of finding the coefficient 
for liquids, because they must be enclosed in some solid, which 
also expands by heat. Hence, the apparent expansion must be 
corrected by allowing for the expansion of the inclosing solid, be 
fore the coefficient of absolute expansion is known. 

314 HEAT. 

This fact is illustrated by the following experiment. Fill the 
bulb and part of the stem of a large thermometer tube with a col 
ored liquid, and then plunge the bulb quickly into hot water ; the 
first effect is, that the liquid falls, as if it were cooled ; after a mo 
ment it begins to rise, and continues to do so till it attains the 
temperature of the hot water. The first movement is caused by 
the expansion of the glass, which is heated so as to enlarge its 
capacity and let down the liquid before the heat has penetrated 
the latter. It is obvious that what is rendered visible in this case, 
must always be true when a liquid is heated namely, that the 
vessel itself is enlarged, and therefore that the rise of the liquid 
shows only the difference of the two expansions. Ingenious 
methods have been devised for obtaining the coefficients of abso 
lute expansion of liquids, and the results are to be found in tables 
on this subject. 

521. Exceptional Case. There is a very important excep 
tion to the general law of expansion by heat and contraction by 
cold, in the case of water just above the freezing point. If Avater 
be cooled down from its boiling point, it continually contracts till 
it reaches a point somewhat above 39 F., when it begins to ex 
pand, and continues to expand till it freezes at 32 F. On the 
other hand, if water at 32 F. be heated, it contracts till it readies 
a point between 39 and 40 F., when it commences to expand. 
Therefore the density of water is greatest at the point where this 
change occurs. Different experimenters vary a little as to its ex 
act place, but it is usually called 4 C., or 39.2 F. 

The importance of this exception is seen in the fact that ice 
forms on the surface of water, and continues to float until it is 
again dissolved. As the cold of winter comes on, the upper stra 
tum of a lake grows more dense and sinks ; and this process con* 
tinues till the temperature of the surface reaches 39, when it is 
arrested. Below that point the surface grows lighter as it becomes 
colder, till ice is formed, and shields the water beneath from the 
severe cold of the air above. 

As in solids so in liquids, the thermal force is very great. Sup 
pose mercury to be expanded by raising its temperature one de 
gree, it would require more than 300 pounds to the square inch to 
compress it to its former volume. 

522. Expansion of Gases. The gases expand by heat more 
rapidly and more regularly than solids and liquids. The large ex 
pansion and contraction of air is made visible by immersing the 
open end of a large thermometer tube in colored liquid. When the 
bulb is warmed, bubbles of air are forced out and rise to the top 


of the liquid ; when it is cooled, the air contracts and the liquid 
rises rapidly in the tube. 

The coefficient of expansion for air is about 0.00205, which in 
creases slightly with increase of temperature and of pressure. 
And- most of the gases have coefficients which differ but little 
from this. 



523. Heat is Communicated in Several Ways. 1. By 

radiation. Heat is said to be radiated when the vibratory motion 
is transmitted from the source with great swiftness through the 
ether which fills space. Its velocity is supposed to be the same as 
that of light. The motion is propagated in straight lines in every 
direction, and each line is called a ray of heat. We feel the rays 
of heat from the sun or a fire, when no object intervenes between 
it and ourselves. 

2. By reflection. When rays of heat, on striking a surface, are 
thrown back from it, they are said to be reflected; and the law of 
reflection is the same as for sound, namely, the angle of incidence 
equals the angle of reflection, and they lie on opposite sides of the 
perpendicular to the surface. 

3. By conduction. This is the slow progress of the vibratory 
motion from one atom to another of ordinary matter. 

4. By convection.^ This mode of communication takes place 
onlj,m fluids. When the particles are expanded by heat, they are 
pressed upward by others which are colder and therefore specifi 
cally heavier. Heat is thus conveyed from place to place by the 
motion of the heated matter. 

524. Radiation cf Heat. The intensity of heat radiated 
from a given kind of source, is governed by the three following 

1. Tlie intensity of radiated heat varies as the temperature of 
the source. 

2. It varies inversely as the square of the distance. 

3. It grows less, while the inclination of the rays to the surface 
of the radiant grows less. 

The truth of these laws is ascertained by a series of careful ex 
periments. But the second may be proved mathematically from 

316 HEAT. 

the fact of propagation in straight lines, and is true of other ema 
nations, such as sound and light. For the heat, as it advances in 
every direction from the radiant, is spread over spherical surfaces 
which increase as the squares of the distances ; therefore the in 
tensities must grow less in the same ratio ; that is, the intensities 
vary inversely as the squares of the distances. 

TJie radiating power of a given body depends on the condition 
of its surface. 

If a cubical vessel filled with hot water have one of its vertical 
sides coated with lamp black, another with mica, a third with tar 
nished lead, and the fourth with polished silver, and the heat ra 
diated from these several sides be concentrated upon a thermome 
ter bulb, the ratio of radiation will be found nearly as follows 

Lamp black, 100 

Mica, 80 

Tarnished lead, 45 

Polished silver, 12 

Polished metals generally radiate feebly; and this explains the 
familiar fact that hot liquids retain their temperature much better 
in bright metallic vessels than in dark or tarnished ones. 

525. Equalization of Temperature. Eadiation is going 
on continually from all bodies, more rapidly in general from those 
most heated ; and therefore there is a constant tendency toward 
an equal temperature in all bodies. A system of exchange goes 
on, by which the hotter bodies grow cool, and the colder ones grow 
warm, till the temperature of all is the same. But this equality 
does not check the radiation ; it still goes forward, each body im 
parting to others as much heat as it receives from them. 

526. Reflection of Heat. "When rays of heat meet the .sur 
face of a body, some of them are reflected, passing off at the same 
angle with the perpendicular on the opposite side. But others 
pass into the body, and are said to be absorbed by it. It is true 
of waves of heat as of all other kinds of vibration, that when they 
meet a new surface and are reflected, the angle of incidence equals 
the angle of reflection, and that their intensity after reflection is 

If a person, when near a fire, holds a sheet of bright tin so as 
to see the light of the fire reflected by it, he will plainly perceive 
that heat is reflected also. And if any sound is produced by the 
fire, as the crackling of combustion, or the hissing of steam from 
wood, the reflection of the sound is likewise heard. This simple 
experiment proves that waves of sound, of heat, and of light, all 
follow the same law of reflection. 



527. Her.t Concentrated by Reflection. Lot two pol 
ished reflectors, 31 and N (Fig. 289), having the form of concave 
paraboloids, be placed ten or fifteen feet apart, with their axes in 
the same straight line, and let a red-hot iron ball be in the focus A 

FIG. 289. 

of one, and an inflammable substance, as phosphorus, in the focus 
I> of the other ; then the latter will be set on fire by the heat of 
the ball. The rays diverging from A to J/are reflected in parallel 
lines to N, and then converged to B. 

If, instead of phosphorus, the bulb of a thermometer is put in 
the focus B y a high temperature is of course indicated on the scale. 
Xow remove the hot ball from A, and put in its place a lump of 
ice ; then the thermometer at B sinks far below the temperature 
of the room. This last experiment does not prove that cold is re 
flected as well as heat, but confirms what was stated (Art. 525), 
that all objects radiate to one another till their temperatures are 
equalized. The ice radiates only a little heat, which is reflected to 
the thermometer, but the latter radiates much more, which is re 
flected to the ice, so that the temperature of the thermometer 
rapidly sinks. 

528. Absorption of Heat. So much of the radiant heat as 
falls on a body and is not reflected, is absorbed. The absorbing 
power in a body is found to be in general equal to its radiating 
power. It is very noticeable that bodies equally exposed to the 
radiant heat of the sun or a fire, become very unequally heated. 
A white cloth on the snow, under the sunshine, remains at the 

318 HEAT. 

surface ; a black cloth sinks, because it absorbs heat, and melts the 
snow beneath it. Polished brass before a fire remains cold ; dark, 
unpolished iron, is soon hot. 

529. Conduction of Heat by Solids. While radiated and 
reflected heat moves through the empty spaces of the solar system, 
and through the atmospheres of the planets, with inconceivable 
velocity, conducted heat, on the contrary, passes through bodies 
very slowly, and yet at very different rates in different bodies. 
Those in which heat is conducted most rapidly, are called good 
conductors, as the common metals ; those in which it passes slowly, 
are called poor conductors, as glass and wood. In general, the 
bodies which are good conductors of heat, are also good conduc 
tors of electricity. Let rods of different metals and other sub 
stances, A, B, C, &c. (Fig. 290), all of the same length, be inserted 
with water-tight joints in the 

side of a wooden vessel. Then ^ IG< 290. 

attach by wax a marble under 
the end of each rod, and fill 
the vessel with boiling water. 
The marbles will fall by the 
melting of the wax, not at the 
same, but at different times, 
showing that the heat reaches 
some of them sooner than oth 
ers. It will be seen, however, in the chapter on specific heat, that 
the order in which they fall is not necessarily the order of con 
ducting power. 

530. Effects of Molecular Arrangement. Organic sub 
stances usually conduct heat poorly ; and bodies having a struc 
tural arrangement which differs in different directions, are not 
likely to conduct equally well in all directions. Thus, let two thin 
plates be cut from the same crystal, one, A (Fig. 291), perpendic- 

FIG. 291. 

ular, and the other, B, parallel to the optic axis. Let a hole be 
drilled through the centre of each, and after a lamina of wax has 
been spread over the crystal, let a hot wire be inserted in it. On 


the plate A, the melting of the wax will advance in a circle, show 
ing equal conducting power in all directions in the transverse sec- 
lion. In the plate B, it will advance in an elliptical form, the 
major axis being parallel to the optic axis of the crystal, proving 
the best conduction to be in that direction. 

A block of wood cut from one side of the trunk of a tree, con 
ducts most perfectly in the direction of the fiber, and least in a 
direction which is tangent to the annual rings and perpendicular 
to the liber, and in an intermediate degree in the direction of the 
radius of the rings. 

531. Conduction by Fluids. Fluids, both liquid and gas 
eous, are in general very poor conductors. Water, for example, 
can be made to boil at the top of a vessel, while a cake of ice is 
fastened within it a few inches below the surface. If ther 
mometers are placed at different depths, while the water boils at 
the top, there is discovered to be a very slight conduction of 
heat downward. The gases conduct even more imperfectly than 

It will be seen hereafter (Art. 533) that a mass of fluid becomes 
heated by convection, not by conduction. 

32. Illustrations of Difference in Conductive Power. 

In a room where all articles are of equal temperature, some feel 
much colder than others, simply because they conduct the heat 
from the hand more rapidly ; painted wood feels colder than 
woolen cloth, and marble colder still. If the temperature were 
higher than that of the blood, then the marble would seem the 
hottest, and the cloth the coolest, because of the same difference 
of conduction to the hand. 

Our clothing does not impart warmth to us, but by its non-con 
ducting property, prevents the vital warmth from being wasted by 
radiation or conduction. If the air were hotter than our blood, 
the same clothing would serve to keep us cool. 

A pitcher of water can be kept .cool much longer in a hot day, 
if wrapped in a few thicknesses of cloth ; for these prevent the 
heat of the air from being conducted to the water. In the same 
way ice may be prevented from melting rapidly. 

The vibrations of heat, like those of sound, are greatly inter 
rupted in their progress by want of continuity in the material. 
Any substance is rendered a much poorer conductor by being in 
the condition of a powder or fiber. Ashes, sand, sawdust, wool, 
fur, hair ; &c., owe much of their non-conducting quality to the 
innumerable surfaces which heat must meet with in being trans 
mitted through them. 



FIG. 292. 

533. Convection of Heat. Liquids and gases are heated 
almost entirely by convection. As heat is applied to the sides and 
bottom of a vessel of water, the heated particles become specifically 
lighter, and are crowded up by heavier ones which take their place. 
There is thus a constant circulation going on which tends to 
equalize the temperature of the whole. This motion is made visi 
ble in a glass vessel, by putting into the 

water some opaque powder of nearly the 
same density as water. Ascending cur 
rents are seen over the part most heated, 
and descending currents in the parts far 
thest from the heat, as represented in Fig. 
292. The ocean has perpetual currents 
caused in a similar manner. The hottest 
portions flow away from the tropical to 
ward the polar latitudes, while at greater 
depths the cold waters of high latitudes 
flow back towards the tropics. 

For a like reason, the air is constantly 
in motion. The atmospheric currents 
on the earth have been considered in Chap 
ter III of Pneumatics. 

534. Diathermancy. It has already been noticed, that ra 
diant heat passes freely through the atmosphere as well as through 
vacant space. The air is therefore said to be diathermal ; it is 
also transparent, since it permits light to pass freely through it. 
But there are substances which allow the free transmission of the 
waves of light, but not those of heat ; and there are others through 
which waves of heat can freely pass, but not those of light. 

Water and glass, which are almost perfectly transparent to the 
faintest light, will not transmit the vibrations of heat unless they 
are very intense. If an open lamp-flame shines upon a thin film 
of ice, while nearly the whole of the light is transmitted, only 6 
per cent, of the heat can pass through. On the other hand, rock 
salt is remarkably diathermal. A plate of it, one-tenth of an inch 
thick, will transmit 92 per cent, of the heat of a lamp ; and if iif 
be coated with lampblack so thick as to stop light completely, the 
heat is still transmitted with almost no diminution. 

If a prism be made of a substance highly diathermal, as rock 
salt, it is found that heat, as well as light, is refracted, being bent 
from its course less than most of the colors, arid falling mostly be 
yond the red extremity of the visible spectrum, but partially coin 
ciding with that color. 




535. Specific Heat. The heat which is absorhed by a body 
is not wholly employed in raising its temperature. While a part 
of the thermal force which is communicated, throws the atoms 
into vibration, that is, heats the body, another part performs inte 
rior work of some other kind, such as urging the atoms asunder, 
or forcing them into new arrangements. This latter portion is 
lost to our sense and to the thermometer, until the body is again 
cooled, when it re-appears. The relative quantity of the force thus 
hidden from view is different in different substances. Hence the 
phrase, specific heat, is used to express the amount of heat required 
to raise a given weight one degree of temperature. The specific 
heat of water is greater than that of any other substance known, 
and it is made the standard of comparison. 

The thermal unit is the amount of heat required to raise the 
temperature of a pound of water one degree, and is called 1. The 
specific heat of a few substances is given in the following table, in 
order to show how greatly they differ. 

Water i.oooo j Silver 0.0570 

Sulphur 0.2026 I Mercury .... 0.0333 

Iron 0.1138 | Gold 0.0324 

Copper . . . . . 0.0951 I Lead 0.0314 

If a pound of water, a pound of iron, and a pound of mercury 
are each raised one degree in temperature, the water consumes 
about nine times as much heat as the iron, and thirty times as 
much as the mercury. 

When bodies are cooled, they show the same differences in the 
quantity of heat which they give off. 

It is a benevolent provision in nature, that water, which is ex 
tended over so large a portion of the globe, has so great specific 
heat ; for the changes of both heat and cold are by this means 
greatly moderated. 

536. Method of Finding Specific Heat. The following 
is one of .several methods of finding the specific heat of a sub 
stance ; it is called the method of mixtures. Let a known weight 
of the substance be heated to a certain temperature, and then 
plunged into, or mixed with, the same weight of water of a low 
temperature ; after which measure the temperature of the mass. 
It will thus be known how much one has lost and the other 


322 HEAT. 

gained in order to reach the common point. If a pound of mer 
cury at the temperature of 132, be poured into a pound of water 
afc the temperature of 32, the mass will be found at about 35.25 C , 
the water being heated only 3.25, while the mercury is cooled 
96.75. Therefore 96.75 : 3.25 : : 1 : 0.0335, which is about the 
specific heat of mercury. 

The specific heat of bodies is in general a little greater as their 
temperature rises. That of the gases, however, seems to be nearly 
constant at all temperatures, and under all pressures. 

537. Apparent Conduction Affected by Specific Heat. 

The conducting power of different substances cannot be correctly 
compared, without making allowance for their specific heat (Art. 
529). For the heat which is communicated to one end of a rod, 
will reach the other end more slowly, if a great share of it disap 
pears on the way. For instance, at the same distance from the 
source of heat, wax is melted quicker on a rod of bismuth than 
on one of iron, though iron is the best conductor, because the 
specific heat of iron is three times as great as that of bismuth ; 
the heat actually reaches the wax soonest through the iron, but 
not enough to melt it, because so much is required to raise the 
iron to a given temperature. 

538. Changes of Condition. Among the most important 
effects produced by heat, are the changes of condition from solid 
to liquid and from liquid to gas, or the reverse, according as the 
temperature of a body is raised or lowered. Increase of heat** 
changes ice to water, and water to steam, and the diminution of 
heat reverses these effects. A large part of the simple substances, 
and of compound ones not decomposed by heat, undergo similar 
changes at some temperature or other ; and probably it would be 
found true of all if the requisite temperature could be reached. 

The melting point (called also freezing point, or point of conge 
lation) ot a substance is the temperature at which it changes from 
a solid to a liquid or the reverse. 

The boiling point is the temperature at which it changes from 
a liquid to a gas or the reverse. 

539. Latent Hsat. Whenever a solid becomes a liquid, or a 
liquid becomes a gas, a large amount of heat disappears, and is 
said to become latent. The thermal force is expended in sunder 
ing the atoms, and perhaps in putting them into new relations 
and combinations, so that there is not the slightest increase of 
temperature after the change begins till it ends. The force is not 
lost, but is treasured up in the form of potential energy, which be- 


comes available whenever a change is made in the opposite direc 
tion. Using the force of heat to turn water into steam, is like 
using the strength of the arm in coiling up a spring, or lifting a 
weight from the earth. The spring and the weight are each in a 
condition to perform work. They have potential energy, which 
can be used at pleasure. 

It has been already noticed that much heat disappears in 
bodies of great specific heat, as their temperature rises. But the 
amount which becomes latent, while a change of condition takes 
place, is vastly greater. Let heat be applied at a uniform rate to 
a mass of water at the temperature of 32, until it rises to the 
boiling point, 212, and note the time occupied. Continuing the 
same uniform supply, it will require 5J times as long to change it 
all into steam. In other words, 180 of heat will raise water from 
the freezing to the boiling point, and (180 x 5| =) 967.1 are re 
quired to change the same into steam, which still remains at the 
temperature of 212 ; the whole of the 967 of thermal force have 
been consumed in the internal work of re-arranging the atoms. 

540. Temperature of Change of Condition. Different 
substances change their condition at very different temperatures. 
Water solidifies at 32 F., mercury at 39, tin at 455, gold at 
201G. Water boils at 212 F., ether at 95, alcohol at 173, mer 
cury at 662. There are some solids, which soften gradually, and 
pass through a large range of temperature before becoming liquid, 
as iron and glass. No definite melting point can be given for such 

Again, many liquids pass into the gaseous state by a slow and 
almost insensible process which goes on at the surface. This is 
called evaporation; and it takes place at all temperatures, but 
more rapidly as the temperature is higher. Even solids evaporate 
without passing through the liquid form. For example, a thin 
film of ice on a pavement w r astes away in cold weather without 

541. Boiling under Pressure. The boiling point for water 
is given as 212 F. This means that water boils at that point 
under the ordinary pressure of the air, and at or near the sea level. 
At that temperature the steam formed has a tension or expansive 
force equal to the atmospheric pressure. But if the pressure were 
diminished, water would boil at a lower temperature. On high 
mountains, boiling water is from 20 to 30 lower in temperature 
than at the ocean level. And under the receiver of an air pump, 
as pressure is gradually taken off, water boils at lower and lower 
points of temperature down to 72. 

324 HEAT. 

The effect of diminished pressure to lower the boiling point is 
well shown by the following familiar experiment : In a thin glass 
flask, boil a little water, and after removing it from the fire, cork 
and invert the flask. The steam which is formed will soon press 
so strongly upon the water as to stop the boiling. When this 
happens, pour a little cold water upon the flask; the water within 
will immediately commence boiling violently, because the vapor is 
condensed and the pressure removed. This effect may be repro 
duced several times before the water in the flask is too cool to boil 
in a vacuum. 

542. Freezing Produced by Melting. Since a great 
amount of heat disappears in a substance as it passes from the 
solid to the liquid state, the loss thus occasioned may produce 
freezing in a contiguous body. When salt and powdered ice are 
mixed, their union causes liquefaction. And if this mixture is 
surrounded by bad conductors, and a tin vessel containing some 
liquid be placed in the midst of it, the latter is frozen by the ab 
straction of heat from it, by the melting of the ice and salt. In 
this way ice creams and similar luxuries are easily prepared in hot 
as well as in cold weather. 

543. Freezing by Evaporation. In like manner, freezing 
by evaporation is explained. Put a little water in a shallow dish 
of thin glass, and set it on a slender wire-support under the re 
ceiver of an air pump. Beneath the wire-support place a broad 
dish containing sulphuric acid. When the air is exhausted, the 
water in a few moments is found frozen. As the pressure of the 
air is taken off, evaporation proceeds with increased rapidity, and 
the requisite heat for this change of condition can be taken only 
from the disli of water. But the atmosphere of vapor retards the 
process by its pressure ; hence the sulphuric acid is placed in the 
receiver, so as to seize upon the vapor as fast as formed, and thus 
render the vacuum more complete. The water is frozen by giving 
up its heat to become latent in the vapor, so rapidly formed ; but 
when this vapor becomes liquid again in combining with the acid, 
the same heat reappears in raising the temperature of the acid. 

Thin cakes of ice may sometimes be procured, even in the hot 
test climates, by the evaporation of water in broad shallow pans 
under the open sky, where radiation by night aids in reducing 
the temperature. The pans should be so situated as to receive the 
least possible heat by conduction. 

544. Spheroidal Condition. When a little water is placed 
in a red-hot metallic cup, instead of boiling violently, and disap 
pearing in a moment, as might be expected, it rolls about quietly 

S T E A M . 


in the shape of an oblate spheroid, and wastes very slowly. So 
drops of water, falling on the horizontal surface of a very hot stove, 
are not thrown off in steam and spray with a loud hissing sound, 
as they are when the stove is only moderately heated, but roll over 
the surface in balls, slowly diminishing in size till they disappear. 
In such cases, the water is said to be in the spheroidal slate. 
Not being in contact with the metal, it assumes the shape of an 
oblate spheroid, in obedience to its own molecular attractions and 
the force of gravity, as small masses of mercury do on a table. 
The reason why the water does not touch the hot metal is, that 
the heat causes a coat of vapor to be instantly formed about the 
drop, on which it rests as on an elastic cushion ; and as the vapor 
is a poor conductor of heat, further evaporation procee4s very 
slowly. It is easily seen that the spheroid does not touch the 
metal, by so arranging the experiment that a beam of light may 
shine horizontally upon the drop, and cast its shadow completely 
separated from that of the hot plate below it, as in Fig. 293. 

FIG. 293. 

If the heated surface is cooling, the temperature may become 
so low that the drop at length touches it, when in an instant vio 
lent ebullition takes place, and the water quickly disappears in 



545. Thermal Force in Steam. It has been already noticed 
that while water is heated, and especially while it is converted into 
steam by boiling, the heat apparently lost is so much force treas 
ured up ready for use, as truly as when strength is expendc-d in 
lifting great weights, which by their descent can do the work de- 



sired. In modern engineering, the force of steam is employed 
more extensively, and for more varied purposes, than any other. 
Every steam-engine is a machine for transforming the internal 
motion of heated steam into some of the visible forms of motion. 

546. Tension cf Steam. When steam is formed by boiling 
water in the open air, its tension is equal to that of the air, and 
therefore ordinarily about fifteen pounds to the square inch. But 
when it is formed in a tight vessel, so that it cannot expand, as 
the temperature of the water is raised the tension is increased in 
a much greater ratio ; because the same steam has greater tension 
at a higher temperature, and besides this, new steam is continually 
added. The following table gives the temperature for successive 
atmospheres of tension: 

I . . 

Degrees of 

. 212 

II . . . 

Degrees of 

. 367 

2 .... 

. 2CI 


. ^74. 

. 27C 

. ^81 

. 204. 

. 187 

r . 

. 3O7 



. 32O 

16 . . . 

. "?00 

7 . 

. ^2 


. 404. 

8 . . . . 

18 . . . 

. 4OQ 

. 3<\ I 


. 414 

10 . 

. .^Q 


. 4l8 

It is seen by the above table that thirty-nine degrees of heat 
are needed to add the second atmosphere of tension, and that the 
number diminishes constantly, so that only four degrees are re 
quired to add the twentieth atmosphere. 

In the formation of ordinary steam at 212, one cubic inch of 
water is expanded to about 1,700 cubic inches of steam, or nearly 
a cubic foot. At higher temperatures, the volume diminishes 
nearly as fast as the temperature increases. 

547. The Steam-engines of Savery and Newcomen. 

The only steam-engines that were at all successful before the great 
improvements made by Watt, were the engine of Savery and that 
of Newcomen. No other purpose was proposed by either than that 
of removing water from mines. 

In the engine of Savery, steam was made to raise water by act 
ing on it directly, and not through the intervention of machinery. 
First, the steam in a vessel was condensed by cold water flowing 
over the outside, and the atmosphere raised water into the ex 
hausted vessel by its pressure. In the next place, steam was let 


into the vessel, and by its tension forced the water out, and raised 
it still higher. The water raised by each part of this operation 
was prevented from returning by a valve, as in a forcing pump. 

Newcomen employed steam in a very different way, namely, as 
a power to work a common pump. The pump rod was attached 
to one end of a working beam, and to the other end of the same 
was attached the rod of the steam piston, which moved steam-tight 
in a cylinder. The end of the beam next to the pump was made 
heavy enough to keep the steam piston at the top of the cylinder, 
when no force was applied. The space beneath the piston being 
filled with steam, a little cold water was injected, the steam con 
densed, and the piston forced down by the weight of the air on 
the top of it. Then, as soon as steam was admitted again below, 
though having no greater tension than the atmosphere, the piston 
was drawn up by the weight of the opposite end of the beam. Since 
the water was raised directly by the weight of the atmosphere, 
after the steam had given it opportunity to act, this invention of 
Newcomen was called the atmospheric engine. 

But in neither of these methods was steam used economically 
as a power. The movements in both cases were sluggish, and a 
large part of the force was wasted, because the steam was com 
pelled to act upon a cold surface, which condensed it before its 
work was done. 

548. The Steam-engine of Watt. Steam did not give 
promise of being essentially useful as a power till Watt, in the 
year 17GO, made a change in the atmospheric engine, which pre 
vented the great waste of force. Newcomen introduced the cold 
water which was to condense the steam into the steam cylinder 
itself; and the cylinder must be cooled to a temperature below 
100, else there would be steam of low tension to retard the de 
scent of the piston. But when the piston was to be raised, the 
cylinder must be heated again to 212, in order that the admitted 
steam might balance the pressure of the air. 

In the engine of Watt, the steam is condensed in a separate 
vessel called the condenser. The steam cylinder is thus kept at 
the uniform temperature of the steam. In the first form which 
he gave to his engine, he so far copied the atmospheric engine as 
to allow the piston, after being pressed down by steam, to be raised 
again by the load on the opposite end of the great beam, while the 
steam circulates freely below and above the piston. This was 
called the single-acting engine, and might be successfully used for 
the only use to which any steam-engine was as yet applied, 
namely, pumping water from mines. But he almost immediately 
introduced the change by which the whole forco of the steam was 



FIG. 294. 

brought to act on the upper and the under side of the piston. It 
thus became double-acting, and the steam force was no longer 

549. The Double-acting Engine. Let 8 (Fig. 294) be the 
steam cylinder, P the piston, A the piston rod, passing with 
steam-tight joint through the top 

of the cylinder, C the condenser, 
kept cold by the water of the cis 
tern G, B the steam pipe from 
the boiler, K the eduction pipe, 
which opens into the valve chest 
at 0, D D the D- valve, E F the 
openings from the valve-chest into 
the cylinder. As the D-valve is 
situated in the figure, the steam 
can pass through B and ^into 
the cylinder below the piston, 
while the steam above the piston 
can escape by F through and K 
to the condenser, where it is con 
densed as fast as it enters ; so that 
in an instant the space above the 
piston is a vacuum, while the 
whole force of the steam is ex 
erted on .the under side. The 

piston is therefore driven upward without any force to oppose it. 
But before it reaches the top, the D-valve, moved by the machin 
ery, begins to descend, and shut off the steam from E and admit it 
to F, and, on the other hand, to shut F from the eduction pipe 0, 
and open E to the same. The steam will then press on the top 
of the piston, and there will be a vacuum below it, so that the pis 
ton descends with the whole force of the steam, and without 
resistance. To render the condensation more sudden, a little cold 
water is thrown into the condenser at each stroke through the 
pipe H. 

550. The Low-pressure Engine. The principle of the low- 
pressure engine is illustrated by the figure and description of the 
preceding article. But the condensing apparatus of this kind of 
engine requires many other parts, most of which are presented in 
Fig. 295. C is the steam cylinder ; R the rod connecting its pis 
ton with the end of the working beam, not represented ; A the 
steam pipe and throttle valve ; B B the D-valve ; D D the educ 
tion pipe, leading from the valve chest to the condenser E\ G G 
the cold water surrounding the condenser; ^the air-pump, which 



keeps tho condenser clear of air, steam, and water of condensation ; 
/the hot well, in which the water of condensation is deposited l.v 
the air-pump; A" the hot-water pump, which forces the water 

FIG. 295. 

in the hot well through L to the boiler ; // the cold-water pump, by 
which water is brought to the cistern G G\ the rods of all the 
pumps, F, K, and If, are moved by the working beam; P the 
fly-wheel ; M the crank of the same, N the connecting-rod, by 
which the working beam conveys motion to the fly-wheel ; Q the 
excentric rod, by which the D-valve is moved; the governor, 
which regulates the throttle valve in the steam pipe A. 

551. Steam- Valves. What are commonly called valves in 
steam machinery are not strictly such, since they are not opened 
by the pressure of a fluid in one direction, and closed by a pres 
sure in the opposite direction. On the contrary, they are opened 
and shut by the action of an eccentric cam on the principal axis. 
The puppet valve is the frustum of a cone, fitting into a conical 
socket, and opens the pipe by being raised. The sliding valve does 
not rise, but slides over the aperture. The rotary valve, like the 
common stop-cock, is a cylinder lying^ across the pipe, and having 
an aperture through it, so that by a quarter revolution it opens or 
shuts the pipe. The throttle valve, like the damper of a stove-pipe, 
is a partition in the pipe, turning on an axis, so as to lie crosswise 
or lengthwise. The D-valve, so called from its form, is a sliding 
valve which takes the place of four valves in the earlier engines, 

330 HEAT. 

connecting both the top and the bottom of the steam cylinder 
with the boiler and with the condenser. 

There is much economy of fuel and saving of wear in the 
machinery, arising from the proper adjustment of the valves. If 
the steam enters the cylinder during the whole length of a stroke 
of the piston, its motion is accelerated ; and is therefore swiftest 
at the instant before being stopped ; thus the machinery receives 
a violent shock. If the valve is adjusted to cut off the steam when 
the piston has made one-third or one-half of its stroke, the dimin 
ishing tension may exert about force enough, during the remain 
ing part, to keep up a uniform motion. The cut-off, however, 
should be regulated in each engine, according to friction and 
other obstructions. 

552. High-Pressure Engine. The engine of Watt, already 
described, is properly called a low-pressure engine, because the 
steam, having a vacuum on the opposite side of the piston, works 
at the least possible tension. For many purposes, especially those 
of locomotion, it is advantageous to dispense with the large weight 
and bulk of machinery necessary for condensation, and do the 
work with steam of a higher tension. In Fig. 295, if the con 
denser, cistern, and all the pumps are removed, then the steam is 
discharged from E and F at each stroke into the air. Therefore 
the steam in that part of the cylinder which is open to the air, 
will have a tension of 15 Ibs. per inch ; and, consequently, the 
steam on the opposite side of the piston must have a tension 15 
Ibs. per inch greater than before, in order to do the same work. 

553. Applications of Steam Power. For more than half 
a centuiy, the only use of the steam-engine was to work the water 
pumps of the English mines. But the genius of Watt has ren 
dered it available for nearly every purpose which requires the use 
of machinery. Every description of machine, for the heaviest and 
the lightest operation, may have a steam-engine for its prime 
mover. Near the beginning of the present century, it began to be 
used for locomotion on land and water ; and at the present day, 
both traveling and the transportation of merchandise are princi 
pally accomplished by means of steam. 

554. Estimation of Steam Power. It is customary to ex 
press the power of a steam-engine by comparing it with the num 
ber of horses whose strength it equals. In making this compari 
son, Watt took as a measure of one horse-power, the ability to raise 
2,000,000 Ibs. through the height of one foot in an hour ; or 2,000,- 
000 foot-pounds per hour. It is obviously immaterial what the 
respective factors for feet, and for pounds, are, if the product only 


equals 2,000,000. For example, 2,000 Ibs. through 1,000 feet, 
or 5,000 Ibs. 400 feet per hour, &c., is equal to one horse-power. 
It is found that the available force of one cubic fool of water, when 
changed to steam, is about equal to 2,000,000 foot-pounds ; that 
13, to one horse-power. Hence, an engine of fifty horse-power is 
one which can change fifty cubic feet of water into steam in one 

555. Mechanical Equivalent of Heat In all cases in 

which mechanical force produces heat, and again in all those in 
which heat produces visible motion, careful experiment proves 
that heat and mechanical force may each be made a measure of 
the other. Forces of any kind may be compared, by observing 
the weights which they will lift through a given distance. The 
mechanical equivalent of heat (commonly called, from the name of 
an English experimenter, Joule s equivalent) is given in the fol 
lowing statement : 

The force required to heat one pound of water one degree P., is 
equal to that which would lift 112 pounds the distance of one foot, 
or is equal to 772 foot-pounds. . 

The force requisite to raise one pound of water 1 F., is some 
times called the thermal unit (Art. 535), and all forces may be 
brought to this as a standard of comparison. Thus, one horse 
power (2,000,000 foot-pounds per hour) is 2,590 thermal units per 
hour, or about 43 per minute. 

Since a force of 772 foot-pounds is expended in heating a 
pound of water 1 F., therefore to heat the same from 32 to 212 
requires a force of 138,900 foot-pounds ; and to change the same 
pound of water into steam of atmospheric tension requires an ad 
ditional force of 746,900 foot-pounds (Art. 539). 



556. Manner in -which the Air is Warmed. The space 
through which the earth moves around the sun is intensely cold, 
probably 75 below zero. And the one or two hundred miles of 
height occupied by the atmosphere is too cold for animal or vege 
table life, except the lowest stratum, three or four miles in thick 
ness. This portion receives its heat mainly by convection. The 

332 HEAT. 

radiated heat of tlie sun passes through tho air, warming it but 
little, and on reaching the earth is partly absorbed by it. The air 
lying in contact with the earth, and thus becoming warmed, grows 
lighter and rises, while colder portions descend and are warmed in 
their turn. So long as the sun is shining on a given region of the 
earth, this circulation is going on continually. But the heated 
air which rises is expanded by diminished pressure, and thus 
cooled. Hence the circulation is limited to a very few miles next 
to the earth. 

557. Limit of Perpetual Frost. At a moderate elevation, 
even in the hottest climate, the temperature of the air is always 
as low as the freezing point. Hence the permanent snow on the 
higher mountains in all climates. The limit at the equator is 
about three miles high, and with many local exceptions it de 
scends each way to the polar regions, where it is very near the 
earth. The descent is more rapid in the temperate than in the 
torrid or frigid zones. 

558. Isothermal Lines. These are imaginary lines on each 
hemisphere, through all those points whose mean annual tempera 
ture is the same. At the equator, the mean temperature is about 
82, and it decreases each way toward the poles, but not equally 
on all meridians. Hence the isothermal lines deviate widely from 
parallels of latitude. Their irregularities are due to the difference 
between land and water, in absorbing and communicating heat, 
to the various elevations of land, especially ranges of mountains, 
to ocean currents, &c. In the northern hemisphere, the isother 
mal lines, in passing westward round the earth, generally descend 
toward the equator in crossing the oceans, and ascend again in 
crossing the continents. For example, the isothermal of 50, 
which passes through China on the parallel of 44, ascends in 
crossing the eastern continent, and strikes Brussels, lat. 51 ; and 
then on the Atlantic, descends to Boston, lat. 42, whence it once 
more ascends to the N. W. coast of America. The lowest mean 
temperature in the northern hemisphere is not far from zero, but 
it is not situated at the north pole. Instead of this, there are two 
poles of greatest cold, one on the eastern continent, the other on 
the western, near 20 from the geographical pole. There are indi 
cations, also, of two south poles of maximum cold. 

559. Moisture of the Atmosphere. By the heat of the 
sun all the waters of the earth form above them an atmosphere of 
vapor, or invisible moisture, having more or less extent and ten 
sion, according to several circumstances. Even ice and snow, at 
the lowest temperatures, throw off some vapor. A gaseous body 


diffuses itself by its force of tension, "whether another gas occupies 
the same space or not; that is, the particles of one do nut i-xrr! 
a perceptible attraction or repulsion on those of the other, but 
each is a vacuum to the other, except so far as it obstructs its 
movements. Therefore, at a given temperature, there can exist an 
atmosphere of vapor of the same height and tension, whether 
there is an atmosphere of oxygen and nitrogen or not. Vapor, 
then, is not strictly suspended in the air, or dissolved by it, but 
exists independently. And yet it is by no means always true that 
there is actually the same tension of vapor as there would be if it 
existed alone, because of the time required for the formation of 
vapor, on account of mechanical obstruction presented by the air; 
whereas, if no air existed, the vapor would form almost instantly. 
It is on the same account that water will boil in a vacuum at 72, 
but under the pressure of the air must be heated to 212. 

560. Temperature and Tension of Vapor. The degree 
of tension of vapor forming without obstruction, depends on its 
temperature, but varies far more rapidly ; increasing pretty nearly 
in a geometrical ratio, while the heat increases arithmetically. 
Hence, if vapor should receive its full increment of tension, while 
the thermometer rises 10 degrees from 80 to 90, a vastly greater 
quantity would be added than when it rises 10 degrees from 40 
to 50. On the contrary, if vapor is at its full tension in each 
case, much more water will be precipitated in cooling from 90 to 
80 than from 50 to 40. 

561. Dew-point This is the temperature at which vapor, 
in a given case, is precipitated into water in some of its forms. 
If there was no air, the dew-point would always be the same as 
the existing temperature ; since lowering the temperature in the 
least degree would require a diminished tension or quantity of 
vapor, some must therefore be condensed into water. But in the 
air the tension may not be at its full height, and therefore the 
temperature may need to be reduced several degrees before precip 
itation will take place. A comparison of the temperature with 
the dew-point is one of the methods employed for measuring the 
humidity of the air. 

562. Measure of Vapor. The measure of the vapor exist 
ing at a given time, is expressed by two numbers, one indicating 
its tension, i.e., the height of the column of mercury which it 
will sustain; the other, humidity, i.e., its quantity per cent., as 
compared with the greatest possible amount at that temperature-. 
Thus, tension = 0.6, humidity = 83, signifies that the quantity of 
vapor is sufficient to support six-tenths of an inch of mercury, and 

334 HEAT. 

is 83 hunclredths of the quantity which could exist at that tem 
perature. The greatest tension possible at zero, is 0.04 ; at the 
freezing point, 0.18 ; at 80, 1.0. At the lowest natural tempera 
tures, the maximum tension is doubled every 12 or 14 ; at the 
highest, every 21 or 22. 

533. Hygrometers. This is the name usually given to in 
struments intended for measuring the moisture of the air. But 
the one most used of late years is called the psychromcter, which 
gives indication of the amount of moisture by the degree of cold 
produced in evaporation ; for evaporation is more rapid, and there 
fore the cold occasioned by it the greater, according as the air is 
drier. The psychrometer consists of two thermometers, one hav 
ing its bulb covered with muslin, and moistened before the ob 
servation. The wet-bulb thermometer will ordinarily indicate a 
lower temperature than the dry-bulb ; if, in a given case, they read 
alike, the humidity is 100. The instrument is accompanied by 
tables, giving tension and humidity for any observation. 

564. Precipitations of Moisture. Whenever the air is 
cooled below the dew-point, a part of the vapor is deposited in the 
liquid or solid form. The precipitations occur under various con 
ditions, and receive the following names : dew, frost, fog, cloud, 
rain, mist, hail, sleet, and snow. 

565. Dew. Frost. The deposition called dew takes place on 
the surface of bodies, by which the air is cooled below its dew- 
point. It is at first in the form of very small drops, which unite 
and enlarge as the process goes on. Dew is formed in the even 
ing or night, when the surfaces of bodies exposed to the sky 
become cold by radiation. As soon as their temperature has 
descended to the dew-point, the stratum of air contiguous to them 
deposits moisture, and continues to do so more and more as the 
cold increases. 

Of two bodies in the same situation, that will receive most 
dew which radiates most rapidly. Many vegetable leaves are 
good radiators, and receive much dew. Polished metal is a poor 
radiator, and ordinarily has no dew deposited on it. 

Sometimes, however, good radiators have little dew, because 
they are so situated as to obtain heat nearly as fast as they radiate 
it. Dew is rarely formed on a bed of sand, though it is a good 
radiator, because the upper surface gets heat by conduction from 
the mass below. Dew is not formed on water, because the upper 
stratum sinks and gives place to warmer ones. 

Bodies most exposed to the open sky, other things being equal, 
have most dew precipitated on them. This is owing to the fact, 

CLOUD, 335 

that in such circumstances, they have no return of heat either by 
reflection or radiation. If a body radiates its heat to a building, 
a tree, or a cloud, it also gets some in return, both reflected and 
radiated. Hence, little dew is to be expected in a cloudy night, 
or on objects surrounded by high trees and buildings. 

AVind is unfavorable to the formation of dew, because it min 
gles the strata, and prevents the same mass from resting long 
enough on the cold body to be cooled down to the dew-point. 

When the radiating body is cooled below the freezing point, 
the water deposited takes the solid form in fine crystals, and is 
called frost. Frost will often be found on the best radiators, or 
those exposed to the open sky, when only dew is found elsewhere. 

566. Fog. This form of precipitation consists of very small 
globules of water sustained in the lower strata of the air. Fug- 
occurs most frequently over low grounds and bodies of water, 
where the humidity is likely to be great. If air thus humid mixes 
with air cooled by neighboring land, even of less humidity, there 
will probably be more vapor than can exist at the intermediate 
temperature, for the reason mentioned in Art. 560. The case may 
be illustrated thus. Let two masses of air of equal volumes be 
mixed, the temperature of one being 40, the other CO ; and each 
containing vapor at the highest tension. Then the mixture will 
have the mean temperature of 50, and the vapor of the mixture 
will also be the arithmetical mean between that of the two masses. 
But, according to the law (Art. 560), the vapor can only have a 
tension which is nearly a geometrical mean between the two, and 
that is necessarily lower than the arithmetical mean ; hence the 
excess must be precipitated. If 8 Ibs. of vapor were in one volume 
and 18 Ibs. in the other, an equal volume of the mixture would 
have 4- (8 + 18) 13 Ibs. of moisture ; but at the mean tempera, 
ture of 50, only Vs x 18 = 12 Ibs. could exist as vapor ; there 
fore one pound must be precipitated. And even if one of the 
masses had a humidity somewhat below 100, still some precipita 
tion is likely to take place. 

567. Cloud. The same as fog, except at a greater elevation. 
Air rising from heated places on the earth, and carrying vapor 
with it, is likely to meet with masses much colder than itself, and 
depositions of moisture are therefore likely to take place. Mount 
ain-tops are often capped with clouds, when all around is clear. 
This happens when lower and warmer strata are driven over them, 
and thus cooled below the dew-point. The same air, as it con 
tinues down the other side, takes up its vapor again, and is as 
transparent as it was before ascending. A person on the summit 

336 HEAT. 

perceives a chilly fog driving by him, but the fog was an invisible 
vapor a few minutes before reaching him, and returns to the same 
condition soon after leaving him. The cloud rests on the mount 
ain ; but all the particles which compose it are swiftly crossing 
over. Clouds are often above the limit of perpetual frost ; they 
then consist of crystals of ice. 

568. Classification of Clouds. The aspects of clouds are 
various, and depend in some measure at least on the circumstances 
of their formation. The usual classification is the following : 

1. Cirrus. This cloud is fibrous in its appearance, like hair or 
flax, sometimes straight, sometimes bent, and frequently at one 
end is gathered into a confused heap of fibers. The cirrus is high, 
and often consists of frozen particles, even in summer. 

2. Cumulus. Tins consists of compact rounded heaps, which 
often resemble mountain-tops covered with snow. This form of 
cloud is confined mostly to the summer season ; it usually begins 
to form after the sun rises, and to disappear before it sets, and is 
rarely seen far from land. The cumulus is generally not so high 
as the cirrus. 

3. Stratus. Sheets or stripes of cloud, sometimes overspread 
ing the whole sky, or as a fog covering the surface of the earth or 
water. The stratus is the most common, and usually lies lowest 
in the air. 

4. 5, 6. Cirro-cumulus, cirro-stratus, cumulo-stratus. Inter 
mediate or combined forms. 

7. Nimbus. A cloud, which forms so fast as to fall in rain or 
snow, is called by this name. 

569. Rain, Mist. Whether the precipitated moisture has the 
form of cloud or rain, depends on the rapidity with which precip 
itation takes place. If currents of air are in rapid motion, if the 
temperature of masses, brought into contact by this motion, are 
widely different, and if their humidity is at a high point, the vapor 
will be precipitated so rapidly, that the globules will touch each 
other, and unite into larger drops, which cannot be sustained. 
Globules of fog and cloud, however, are specifically as heavy as 
drops of rain ; but they are sustained by the slightest upward 
movements of the air, because they have a great surface compared 
with their weight. A globule whose diameter is 100 times less 
than that of a drop of rain, meets with 100 times more obstruction 
in descending, since the weight is diminished a million times 
( T Jo) 8 , and the surface only ten thousand times ( T Jo) 2 . So the 
dust of even heavy minerals is sustained in the air for some time, 
when the same substances, in the form of sand, or coarse gravel, 
fall instantly. 


J//W is fine rain ; the drops are barely large enough to make 
their way slowly to the earth. 

570. Hail, Sleet, Snow. When the air in which rapid pre 
cipitation occurs, is so cold as to freeze the drops, hail is produced. 
As hailstones are not usually in the spherical form when they 
reach the earth, it is supposed that they are continually receiving 
irregular accretions in thuir descent through the vapor of the air. 
Hail-storms are most frequent and violent in those regions where 
hot and cold bodies of air are most easily mixed. Such mixtures 
are rarely formed in the torrid zone, since there the cold air is at a 
great elevation ; in the frigid zone, no hot air exists at any height; 
but in the temperate climates, the heated air of the torrid, and the 
intensely cold winds of the frigid zone, may be much more easily 
brought together; and accordingly, in the temperate zones it is 
that hail-storms chiefly occur. Even in these climates, they are 
not frequent except on plains and in valleys contiguous to mount 
ains which are covered with snow during the summer. The 
slopes of the mountain sides give direction to currents of air, so 
that masses of different temperature are readily mingled together. 

Sleet is frozen mist, that is, it consists of very small hailstones. 

Snoiu consists of the small crystals of frozen cloud, united in 
flakes. Like all transparent substances, when in a pulverized 
state, it owes its whiteness to innumerable reflecting surfaces. A 
cloud, when the sun shines upon it, is for the same reason intensely 

571. Theories of Precipitation. It is probable that clouds 
and rain are caused not only by the mixing of air of different tem 
peratures, but also by the changes which take place in the condi 
tion of the air as it ascends. 

In the lower strata, the air is about one degree colder for every 
300 feet of elevation. If, therefore, a mass of air is transferred 
from the surface of the earth to a height in the atmosphere, it will 
be cooled to the temperature of the stratum which it reaches ; not 
principally by giving off its heat, but by expanding, and thus 
having its own heat reduced by being diffused through a larger 
space. Xow, if the rising mass was saturated with moisture, this 
moisture would begin at once to be precipitated by the cooling 
which it undergoes in consequence of expansion. If, instead of 
being saturated, its dew-point is a certain number of degrees below 
its temperature, it must ascend far enough to be cooled to the dew- 
point, before precipitation of its moisture will take place. Sup 
pose, for instance, the temperature at the earth is 70, and the dew- 
point is G5 ; then after the warm air has risen 1500 feet (5 x 300 

338 HEAT. 

ft.), it will become 5 cooler, and contain all the moisture which 
is possible at that temperature. At that point precipitation be 
gins, and forms the base of a cloud. The clouds, called cumulus, 
which are seen forming during many summer forenoons, are the 
precipitations of columns rising from warm spots of earth so high 
that they are cooled below their dew-point. But the movement 
and the precipitation do not stop here ; for, as moisture is precip 
itated, its latent heat is given off in large quantities, which ele 
vates the temperature of the mass, and causes it to rise still higher, 
and precipitate still more of its moisture. As it becomes rarer, 
it spreads laterally, and causes the cumulus often to assume the 
overhanging form which distinguishes that species of cloud. 

572. Cyclones. The late Mr. Kedfield investigated with great 
success the phenomena of violent storms, especially of Atlantic 
hurricanes, and showed that they are generally, if not always, 
great whirlwinds, called cyclones. They usually take their rise in 
the equatorial region eastward of the West India Islands ; they 
rotate on a vertical axis, advancing slowly to the northwest, until 
they approach the coast of the United States near the latitude of 
30, and then gradually veer to the northeast, running nearly par 
allel to the American coast, and finally spend themselves in the 
northern Atlantic. Their rotary motion is always in one direc 
tion, namely, from the east through the north to the west, or 
against the sun. This motion is also far more violent, especially 
in the central parts of the storm, than the progressive motion. 
The rotary motion may amount to 50 or 100 miles per hour, while 
the forward motion of the storm is not more than 15 or 20 miles. 

In the southern hemisphere also, cyclones occur, having a pro 
gressive and a rotary motion, both symmetrical with those of the 
northern cyclones. On the axis they revolve with the sun, not 
against it ; and they first advance toward the southwest, and grad 
ually veer toward the southeast, as they recede from the equator. 

573. Draught of Flues. The effect of the sun s heat in 
causing circulation of the air has been already considered (Art. 
289-293). Similar movements on a limited scale are produced 
whenever a portion of the air is heated by artificial means. Thus, 
the air of a chimney is made lighter by a fire beneath it, than a 
column of the outer air extending to the same height. It is there 
fore pressed upward by the heavier external air, which descends 
and moves toward the place of heat. The difference of weight in 
the two columns is greater, and therefore the draught stronger, if 
the chimney is high, provided the supply of heat is sufficient to 
maintain the requisite temperature. Chimneys are frequently 


built one or two hundred feet high for the uses of manufactories. 
The high fireplaces and large flues of former times were unfavora 
ble for draught, both because much cold air could mingle with 
that which was heated, and because there was room for external 
air to descend by the side of the ascending column. For good 
draught, no air should be allowed to enter the flue except that 
which has passed through the fire. 

574. Ventilation of Apartments. The air of an apart 
ment, as it becomes vitiated by respiration, may generally be re 
moved, and fresh air substituted, by taking advantage of the same 
inequality of weight in air-columns, which has been mentioned. 
If opportunity is given for the warm impure air to escape from the 
top of a room, and for external air to take its place, there will be 
a constant movement through the room, as in the flue of a chim 
ney, though at a slower rate. If the external air is cold, the 
weight of the columns differs more, and therefore the ventilation 
is more easily effected. But in cold weather, the air, before being 
admitted to the room, is warmed by passing through the air-cham 
bers of a furnace. When there is a chimney-flue in the wall of a 
room, with a current of hot air ascending in it, the ventilation is 
best accomplished by admitting the air into the flue at the upper 
part of the room ; since it will then be removed with the velocity 
of the hot-air current. 

The tendency of the air of a warm room to pass out near the 
top, while a new supply enters at the lower part, is shown by hold 
ing the flame of a candle at the top, and then at the bottom, of a 
door which is opened a little distance. The flame bends outward 
at the top and inward at the bottom. 

The impure air of a large audience-room is sometimes removed 
by a mechanical contrivance, as, for instance, a fan-wheel placed 
above an opening at the top, and driven by steam. 

The ventilation of mines is accomplished sometimes by a fire 
built under a shaft, fresh air being supplied by another shaft, and 
sometimes by a fan-wheel at the top of the shaft. If there happen 
to be two shafts which open to the surface at very different eleva 
tions, ventilation may be effected by the inequality of temperature 
which is likely to exist within the earth and above it. Let MM 
(Fig. 296) be the vertical section of a mine through two shafts A 
and B, which open at different heights to the surface of the earth. 
If the external air is of the same temperature as the air within 
the earth, then the column A in the longer shaft has the same 
weight as B and C together, measured upward to the same level. 
In that case, which is likely to occur in spring and fall, there is 
no circulation without the use of other means. But in summer 



the air C is warmer than A and B ; therefore A is heavier than 
B + C. Hence there is a current of air down A and up B. In 
winter, C is colder than air within the earth ; therefore B -f C 
are together heavier 
than A, and the 
current sets in the 
opposite direction, 
down B and up A. 

575. Sources 
of Heat. TJie sun, 
although nearly a 
hundred millions of 
miles from the earth, 
is the source of 
nearly all the heat 
existing at its sur 
face. The interior 
of the earth, except a thickness of forty or fifty miles next to the 
surface, is believed to be in a condition of heat so intense that all 
the materials composing it are in the melted state. But the earth s 
crust is so poor a conductor that only an insensible fraction of all 
this heat reaches the surface. 

Mechanical operations are usually attended by a development 
of heat. For example, if a broad surface of iron were made to re 
volve, rubbing against another surface, nearly all the force ex 
pended in overcoming the friction would appear as heat, a com 
paratively small part being conveyed through the air as sound. 
The cutting tool employed in turning an iron shaft has been known 
to generate heat enough to raise a large quantity of cold water 
to the boiling point, and to keep it boiling for an indefinite time. 
It is a fact familiar to all, that violent friction of bodies against 
each other will set combustibles on fire. The axles of railroad 
cars are made red-hot if not duly oiled ; boats are set on fire by 
the rope drawn swiftly over the edge by a whale after he is har 
pooned ; a stream of sparks flies from the emory wheel when steel 
is polished, &c. Condensation and percussion, as well as friction, 
and all sudden applications of force, cause sensible heat. Indeed, 
wherever the full equivalent of any force is not obtained in some 
other form, the deficiency may be detected in the heat which is 

Chemical action is another very common source of heat. Com 
bustion is the effect of violent chemical attraction between atoms 
of different natures, when both light and heat are manifested. If 
the union goes on slowly, as in the rusting of iron, the amount of 


heat is the same, but it is diffused as fast as developed. The molec 
ular forces, expended in most cases of chemical combination, as 
measured by their heating effects, are enormously great. 

The warmth produced by the vital processes in plants and 
animals is supposed by many physicists to be caused by chemical 
action. In breathing the air, some of its oxygen is consumed, 
which becomes united with the blood. This process is in some 
respects analogous to a slow combustion, by which heat is evolved 
in the animal system. 

L i a HE T . 



576. Definitions. Light is supposed to consist of exceed 
ingly minute and rapid vibrations in a medium or ether which 
fills space ; which vibrations, on reaching the retina of the eye, 
cause vision, as the vibrations of the air cause hearing, when they 
impinge on the tympanum of the ear, and as thermal vibrations 
produce a sensation of warmth, when they fall on the skin. 

Bodies, which of themselves are able to produce vibrations in 
the ether surrounding them, are said to emit light, and are called 
self -luminous, or simply luminous ; those, which only reflect light, 
are called non-luminous. Most bodies are of the latter class. A 
ray of light is a line, along which light is propagated ; a learn is 
made up of many parallel rays ; & pencil is composed of rays either 
diverging or converging ; and is not unfrequently applied to those 
which are parallel. 

A substance, through which light is transmitted, is called a 
medium ; if objects are clearly seen through the medium, it is 
called transparent ; if seen faintly, semi-transparent ; if light is 
discerned through a medium, but not the objects from which it 
comes, it is called translucent; substances which transmit no light 
are called opaque. 

577. Light Moves in Straight Lines. So long as the 
medium continues uniform, the line of each ray is perfectly 
straight. For an object cannot be seen through a bent tube; 
and if three disks have each a small aperture through it, a ray 
cannot pass through the three, except when they are exactly in a 
straight line. The shadow which is projected through space from 
an opaque body proves the same thing; for the edges of the 
shadow, taken in the direction of the rays, are all straight lines. 


From every point of a luminous surface light emanates in all 
possible directions, when not prevented by the interposition of 
an opaque body. Thus, a candle is seen by night at the distance 
of one or two miles ; and within that limit, no space so small as 
the pupil of the eye is destitute of rays from the candle. A 
point from which light emanates is called a radiant. If light 
from a radiant falls perpendicularly on a circular disk, the pencil 
is a cone ; if on a square disk, it is a square pyramid, &c., the 
illuminated surface in each case being the base, and the radiant 
the vertex. 

578. The Velocity of Light It has been ascertained by 
several independent methods, that light moves at the rate of about 
192,500 miles per second. 

One method is by means of the eclipses of Jupiter s satellites. 
The planet Jupiter is attended by four moons which revolve about 
it in short periods. These small bodies are observed, by the tele 
scope, to undergo frequent eclipses by falling into the shadow 
which the planet casts in a direction opposite to the sun. The 
exact moment when the satellite passes into the shadow, or comes 
out of it, is calculated by astronomers. But sometimes the earth 
and Jupiter are on the same side, and sometimes on opposite sides 
of the sun ; consequently, the earth is, in the former case, the 
whole diameter of its orbit, or about one hundred and ninety 
millions of miles nearer to Jupiter than in the latter. Now it is 
found by observation, that an eclipse of one of the satellites is 
seen about sixteen minutes and a half sooner when the earth is 
nearest to Jupiter, than when it is most remote from it, and con 
sequently, the light must occupy this time in passing through the 
diameter of the earth s orbit, and must therefore travel at the rate 
of about 192,000 miles per second. 

Another method of estimating the velocity of light, wholly 
independent of the preceding, is derived from what is called the 
aberration of the fixed stars. The apparent place of a fixed star 
is altered by the motion of its light being combined with the mo 
tion of the earth in its orbit. The place of a luminous object is 
determined by the direction in which its light meets the eye. But 
the direction of the impulse of light on the eye is modified by the 
motion of the observer himself, and the object appears forward of 
its true place. The stars, for this reason, appear slightly displaced 
in the direction in which the earth is moving ; and the velocity 
of the earth being known, that of light may be computed in the 
same manner as we determine one component, when the angles 
and the other component are known. 

The velocity of light has been determined also by direct ex- 

344 LIGHT. 

periment, in a manner somewhat analogous to that employed by 
Wheatstone for ascertaining the velocity of electricity. 

579. Loss of Intensity by Distance. The intensity of 
liglit varies inversely as the square of the distance. In Fig. 297, 
suppose light to radi 
ate from S, through FlG - 2 < 
the rectangle A C 9 
and fall on E G, paral 
lel to A C. A&SAE, 

S B F, &c., are ^Mtr -Ill 

straight lines, the tri- 

are similar, as also the 
rectangles, A C, EG; therefore, A C : E G : : A B* : E F 2 : : SA : 
S E\ But the same quantity of light, being diffused over A C and 
E G, will be more intense, as the surface is smaller. Hence, the 
intensity of light at E : intensity at A : : A C : E G : : S A : S ?, 
which proves the proposition. This demonstration is applicable 
to every kind of emanation in straight lines from a point. 

580. Brightness the Same at all Distances. The bright 
ness of an object is the quantity of light which it sheds, as com 
pared with the apparent area from which it comes. Now the quan 
tity (or intensity), as has just been shown, varies inversely as the 
square of the distance. The apparent area of a given surface also 
diminishes in the same ratio, as we recede from it. Hence, the 
brightness is constant. For illustration, if we remove to three 
times the distance from a luminous body, we receive into the 
eye nine times less light, but the body also appears nine times 
smaller, so that the relation of light to apparent area remains the 

581. Loss of Intensity by Absorption. In a uniform 
medium, while the distance increases arithmetically, the intensity 
diminishes geometrically. Imagine the medium to be divided by 
parallel planes into strata of equal thickness ; and suppose the 

first stratum to diminish the intensity by - of the whole. Then 
the intensity of the light which reaches the second stratum is 

1 - . But on account of the uniformity of the me- 

n n 

dium, every stratum produces the same effect, that is, it transmits 

to the next, of that which falls upon it. Therefore, 

n n 


of --, or ~- 9 leaves the second stratum, * pJl, the t \\l\\\. 

and so on, in a geometrical series. For example, if a piece of 
colored glass is 1] inch thick, and each quarter of an inch u 1 ,- 
sorbs -ij of the light winch falls upon it, then about one-hun 
dredth of what enters the first surface will escape from the last. 

For f?Vs= .01 nearly. 

582. Photometers. These are instruments designed for the 
measurement of the relative intensities of light. We cannot de 
termine by the eye alone how many times more intense one light 
is than another, though we can judge with tolerable accuracy 
when two surfaces are equally illuminated. Photometers are, 
therefore, generally constructed on the plan of determining the 
ratio of intensities of two lights, by means of our ability to decide 
when they illuminate two surfaces equally. It is sufficient to 
mention Kumford s method by shadows. Let the two unequal 
lights be so placed that the two shadows of an opaque body cast 
by them shall fall side by side on a white screen. If one shadow 
appears more luminous than the other, remove to a greater dis 
tance the light which illuminates it (or bring the other nearer), 
until the shadows appear of the same degree of illumination. 
Then measure the distances from the lights to the screen, and the 
intensities of the lights will be directly as the squares of the dis 
tances. For the light at the greater distance, since it illuminates 
the screen equally with the other, must gain as much by intensity 
as it loses by distance ; that is, in the ratio of the square of the 

583. Shadows. When a luminous body shines on one which 
is opaque, the space beyond the latter, from which the light is 
excluded, is called a shadow. The same word, as commonly used, 
denotes only the section of a shadow made by a surface which 
crosses it. Shadows are either total or partial If tangents are 
drawn on all the corresponding sides of the two bodies, the space 
inclosed by them beyond the opaque body is the total shadow; if 
other tangents are drawn, crossing each other between the bodies, 
the space between the total shadow and the latter system of tan 
gents is the partial shadow, or penumbra. In case the bodies are 
spheres, as in Fig. 298, the total shadow will be a cylinder, or con 
ical frustum, each of infinite length, or a complete cone, according 
to the relative size of the spheres. But, in every case, the penum 
bra and inclosed total shadow will form an increasing frustum. 
It is obvious that the shade of the penumbra grows gradually 
deepsr from the outer surface to the total shadow within it. 



Every shadow cast by the sun has a penumbra bordering it, 
which gives to the shadow an ill-defined edge ; and the more re- 

FIG. 298. 

mote the sectional shadow is from the opaque body which casts it, 
the broader will be the partial shadow on the edge. 



584. Radiant and Specular Reflection. Light is said to 
be reflected when, on meeting a surface, it is turned back into the 
same medium. In ordinary cases of reflection, the light is diffused 
in all directions, and it is by means of the light thus scattered 
from a body that it becomes visible, when it sheds no light of its 
own. This is called radiant reflection. It is produced by unpol 
ished surfaces. But when a surface is highly polished, a beam of 
light falling on it is reflected in some particular direction ; and, 
if the eye is placed in this reflected beam, it is not the reflecting 
surface which is seen, but the original object, apparently in a new 
position. This is called specular reflection. It is, however, gene 
rally accompanied by some degree of radiant reflection, since the 
reflector itself is commonly visible in all directions. Ordinary 
mirrors are not suitable for accurate experiments on reflection, 
because light is modified by the glass through which it passes. 
The speculum is therefore used, which is a reflector made of solid 
metal, and accurately ground to any required form, either plane, 
convex, or concave. The word mirror is, however, much used in 
optics for every kind of reflector. 



FIG. 209. 

Optical experiments are usually performed on a beam of light 
admitted through an aperture into a darkened room; the direction 
of the beam being regulated by an adjustable mirror placed out 
side. An instrument consisting of a plane speculum moved by a 
clock, in such a manner that the reflected sunbeam shall remain 
stationary at all hours of the day, is called a heliostat. 

585. The Law of Reflection. When a ray of light is inci 
dent on a mirror, the angle between it and a perpendicular to the 
surface at the point of incidence, is called the angle of incidence; 
and the angle between the reflected ray and the same perpendicu 
lar, is called the angle of reflect ion. The law of reflection found 
to be universally true is the following : 

TJie angles of incidence and reflection are on opposite sides of 
the perpendicular, and are equal to each other. 

This is well shown by attaching a small mirror to the centre 
of a graduated semicircle perpendicular to its plane. Let M D N 
(Fig. 299) be the semicircle, graduated from D both ways to M 
and N 9 and mounted so that it can 
be revolved on its centre, and 
clamped in any position. Let the 
small mirror be at O, with its plane 
perpendicular to CD , then a ray 
from the heliostat, as A C, passing 
the edge at a particular degree, 
will be seen after reflection to pass 
the corresponding degree in the 
other quadrant. By revolving the 
semicircle, any angle of incidence 
may be tried, and the two rays are 
always found to be in the same 
plane with CD, and equally in 
clined to it. 

As the mirror revolves, the re 
flected ray revolves tiuice as fast. 

For A CD is increased or diminished by the angle through 
which the mirror turns; therefore D C B is also increased or 
diminished by the same; hence A C B, the angle between the two 
rays, is increased or diminished by the sum of both, or twice the 
same angle. 

It follows from the law of reflection, that a ray which falls on 
a mirror perpendicularly, retraces its own path after reflection. It 
is obvious, also, that the complements of the angles of incidence 
and reflection are equal, i. e. A C M= B C N. The law of reflec 
tion is applicable to curved as well as to plane mirrors ; the radius 

348 LIGHT. 

of curvature at any point being the perpendicular with which the 
incident and reflected rays make equal angles. 

Radiant reflection forms no exception to the foregoing law, 
though the incident rays are in one and the same direction, and 
the reflected rays are scattered every way. For the minute cavi 
ties and prominences which constitute the roughness of the gene 
ral surface are bounded by small surfaces lying at all inclinations ; 
and each one reflecting the rays which meet it in accordance with 
the law, those rays are necessarily thrown off in all possible 

588. Inclination of Rays to each other not altered by 
the Plane Mirror. 

1. Eays which diverge before reflection, diverge at the same 
angle after reflection. 

Let M N (Fig. 300) be a plane mirror, 
and A >, A C, any two rays of light fall 
ing upon it from the radiant A, and re 
flected in the lines B E, C G. Draw the 
perpendicular A P, and produce it indefi 
nitely, as to F, behind the mirror; also 
produce the reflected rays back of the 
mirror. Let Q R be perpendicular to the 
mirror at the point B\ it is therefore 
parallel to A F, and the plane passing 
through A F and Q R, is that which in 
cludes the ray A B. B E. Therefore, 

E B, when produced back of the mirror, intersects A P produced. 
Let F be the point of intersection. B A F = A B Q, and A FB 
= E B Q-, butAB Q = E B Q (Art, 585); /. B A F=A F B, 
and A B = F B. If P and B be joined, P B being in the plane 
M N is perpendicular to A F, and therefore bisects it. Hence, the 
reflected ray meets the perpendicular A Fas far behind the mir 
ror, as the incident ray does in front. In the same way it may be 
proved that A C= C F, and that C G, when produced back of the 
mirror, meets A F at the same point F. 

Now, since the triangles A C B and FOB, have their sides 
respectively equal, their angles are equal also ; hence B A C 
B F C. Therefore any two rays diverge at the same angle after 
reflection as they did before reflection. 

Since the reflected rays seem to emanate from F, that point is 
called the apparent radiant ; A is the real radiant. 

2. Eays which converge before reflection, converge at the same 
angle after reflection. Let E B, G C, be incident rays converging 
toward F, and let B A, C A, be the reflected rays. It may be 


proved as before, that A and .Fare in the same perpendicular, .1 / , 
and equidistant from P, and that E F G B A C. 

The point F, to which the incident rays were converging, is 
called the virtual focus ; A is the real focus. 

3. Rays which are parallel before reflection are parallel after 

It has been proved in case 1, that F, the intersection of the 
reflected rays, is as far behind the mirror, as A, the intersection of 
incident rays, is before it. Now, if the incident rays are parallel, 
A is at an infinite distance from the mirror. Therefore F is at 
an infinite distance behind it, and the reflected rays are parallel. 

In all cases, therefore, rays reflected by a plane mirror retain 
the same inclination to each other which they had before reflection. 

587. Spherical Mirrors. A spherical mirror is one which 
forms a part of the surface of a sphere, and is either convex or 
concave. The axis of such a mirror is that radius of the sphere 
which passes through the middle of the mirror. In the practical 
use of spherical mirrors, it is found that the light must strike the 
surface very nearly at right angles ; hence, in the following state 
ments, the mirror is supposed to be a very small part of the whole 
spherical surface, and the rays nearly coincident with the axis. 

It is sufficient to trace the course of the rays on one side of the 
axis, since, on account of the symmetry of the mirror around the 
axis, the same effect is produced on every side. 

588. Converging Effect of a Concave Mirror. 

1. Parallel rays are converged to the middle point between the 
centre and surface, which is therefore called the focus of parallel rays, 
or the principal focus. Let R A, L E (Fig. 301), be parallel rays 
incident upon the concave mirror A B, whose centre of concavity is 

FIG. 301. 


C. The ray L E, passing through C, and therefore perpendicular 
to the mirror at E, is reflected directly back. Join C A, and make 
C A F= R A C , then R A is reflected in the line A F, and the 
two reflected rays meet at F. RAC^ACF, .: ACF=FA C, 
and A F C F\ and as A and E are very near together, E F 
F C\ that is, the focus of parallel rays is at the middle point be 
tween C and K 

350 LIGHT. 

2. Diverging rays, fulling on a given concave mirror, are re 
flected converging, parallel, or less diverging, according to the 
degree of divergency in the original pencil. Let C (Fig. 302) be 
the centre of concavity, and F the focus of parallel rays. Then, 

FIG. 302. 

rays diverging from any point, A, beyond (7, will be converged to 
some point, a, between (7 and F, since the angles of incidence and 
reflection are less than those for parallel rays. Eays diverging 
from C are reflected back to C; those from points between C and 
F, as a, are converged to points beyond (7, as A ; those diverging 
from F become parallel ; and those from points between F and the 
mirror, as D, diverge after reflection, but at a less angle than be 
fore, and seem to flow from A 9 . To prove, in the last case, that 
the angle of divergence, A , after reflection, is less than the angle 
D, the divergence before reflection, observe that the angle A is 
less than the exterior angle H B C , but H B C = D B C (Art. 
585), which is less than D B R, which is equal to A D B\ much 
more, then, is A less than A D B. 

3. Converging rays are made to converge more. The rays H B, 
A E, converging to A , are reflected to D, nearer the mirror than 
F is. And it has been shown that the angle D is larger than A , 
hence the convergency is increased. 

From the three foregoing cases, it appears that the concave 
mirror always tends to produce convergency ; since, when it does 
not actually produce it, it diminishes divergency. 

589. Conjugate Foci. When light radiates from A, it is 
reflected to a ; when it radiates from a, it meets at A. Any two 
such interchangeable points are called conjugate foci. If the radius 
of the mirror and the distance of one focus from the mirror are 
given, the distance of its conjugate focus may be determined. 
Let the radius = r ; the distance A E = m ; and a E = n. As 
the angle A B a is bisected by B C, A B : a B : : A C:a C; that 
is, since B E is very small, ~A E : a E : : A C : a (7, or, m : n : : 
m r : r n. 

n r m r 

.*. m = ; and n = = . 

2 n r 2m r 

If A is not on the axis of the mirror, as in Fig. 303, let a line 



FIG. 803. 

be drawn through A and (7, meeting the mirror in E\ this is called 
a secondary axis, and the light radiating from A will be rcilccttd 
to a on the same secondary axis, 
for A E is perpendicular to the 
mirror, and will be reflected di 
rectly back ; and if A E and C E 
are given, a E may be found as 

590. Diverging Effect of a Convex Mirror. 

1. Parallel rays are reflected diverging from the middle point 
between the centre and surface. Let C (Fig. 304) be the centre 
of convexity of the mirror M N, and draw the radii, G M, C D, 

FIG. 304. 

producing them in front of the mirror ; these are perpendicular to 
the surface. The ray R D will be reflected back ; A M will be 
reflected in M B, making B M E A ME. Produce the re 
flected ray back of the mirror, and it will meet the axis in F, mid 
way from C to D-, for FCM=A ME, and FMC=BME\ 
therefore the triangle F C Mis isosceles, and C F = F M, and as 
M is very near D, C F= FD. Hence the rays, after reflection, 
diverge as if they radiated from a point in the middle of C D, 
which is the apparent radiant. 

2. Diverging rays have their divergency increased. Let A D, 
A M(Fig. 305), be the diverging rays; D A, MB, the reflected 

FIG. 305. 

rays ; these when produced meet at F, which is the apparent radi 
ant, MA Fia the divergency of the incident rays, and A F 
of the reflected rays. Now the exterior angle, A F B, is greater 
than C M F, or B M E, or A M E. But A M E, being exterior, 
is greater than M A F; much more, then, is A F B greater than 
MA F. 

3. Convergent rays are at least rendered less convergent, and 

352 LIGHT. 

may become parallel or divergent, according to the degree of pre 
vious convergency. The two first effects are shown by Figs. 304 
and 305, reversing the order of the rays. And it is easy to per 
ceive that rays converging to C, will diverge from C after reflec 
tion ; if to a point more distant than C, they will diverge afterward 
from a point between G and F (Fig. 304), and vice versa. 

The general effect, therefore, of a convex mirror, is to produce 

A and F (Fig. 305) are called conjugate foci, being inter 
changeable points ; for rays from A move after reflection as though 
from F, and rays converging to F are by reflection converged to 
A. Conjugate foci, in the case of the convex mirror, are in the 
same axis either principal or secondary, as they are in the concave 
mirror, and for the same reason, viz., that every axis is perpendic 
ular to the surface. 

591. Images by Reflection. An optical image consists of 
a collection of focal points, from which light either really or appa 
rently radiates. When rays are converged to a focus they do not 
stop, but cross, and diverge again, as if originally emanating from 
the focal point. A collection of such points, arranged in order, 
constitutes a real image. When rays are reflected diverging, they 
proceed as though they emanated from a point behind the mirror. 
A collection of such imaginary radiants forms an apparent or vir 
tual image. The images formed by plane and convex mirrors are 
always apparent; those formed by concave mirrors may be of 
either kind. 

592. Images by a Plane Mirror. When an object is before 
a plane mirror, its image is at the same distance behind it, of the 
same magnitude, and equally inclined to it. Let M N (Fig. 306) 
be a plane mirror, and A B an ob 
ject before it, and let the position Fic - 306. 

of the object be such that the re 
flected rays may enter the eye 
placed at H. From A and B let 
fill upon the plane of the mirror 
the perpendiculars A E, B G, and 
produce them, making E a A E, 
and G b = B G. Now, since the 
rays from A will, after reflection, 

radiate as if from a (Art. 586), and those from B, as if from I, and 
the same of all other points, therefore the image and object are 
equally distant from the mirror. A (7, a c, parallel to the mirror, 
are equal ; as B G = b G, and A E = a E, therefore, by subtrac 
tion, B G = b c ; also the right angles at C and c are equal. There- 


fore A B = a ft, and B A C = I a c\ that is, the object and image 
are of etjiuil size, and equally inclined to the mirror. 

It appears from the demonstration, that the object and its 
image are comprehended between the same perpendiculars to the 
plane of the mirror. 

The object and image obviously have to each other twice the 
inclination that each has to the mirror. Hence, in a mirror in 
clined 45 to the horizon, a horizontal surface appears vertical, 
and one which is vertical appears horizontal. 

593. Symmetry of Object and Image. All the three di 
mensions of the object and image are respectively equal, as shown 
above, but one of them is inverted in position, namely, that dimen 
sion which is perpendicular to the mirror. Hence, a person and 
his image face in opposite directions ; and trees seen in a lake have 
their tops downward. Those dimensions which are parallel to the 
mirror are not inverted. In consequence of the inversion of one 
dimension alone, the object and its image are not similar, but 
symmetrical forms; and one could not coincide with the other if 
brought to occupy the same space. The image of a right hand is 
a left hand, and all relations of right and left are reversed. It is 
for this reason that a printed page, seen in a mirror, is like the 
type with which it was printed. 

594. The Length of Mirror Requisite for Seeing an 
Object. If an object is parallel to a mirror, the length of mirror 
occupied by the image is to the length of the object as the reflected 
ray to the sum of the incident and reflected 

rays. Let A B (Fig. 307) be the length of the 
object, CD that of the image, and F G that of 
the space occupied on the mirror ; then, by 
similar triangles, F O : C D : : EF: E C. But 
CD = A B, and CF= A F; .-. FG : A B : : 
EF : A F + F E. If the eye is brought 
nearer the mirror, the space on the mirror oc 
cupied by the image is diminished, because E F 
has to A F + F E a less ratio than before. The same effect is pro 
duced by removing the object further from the mirror. The length 
of mirror necessary for a person to see himself is equal to half his 
height, because in that case, EF: A F + FE : : 1 : 2, which ratio 
will not be altered by change of distance. 

595. Displacement of Image by Two Reflections. If an 

image is seen by light reflected from two mirrors in a plane per 
pendicular to their common section, its angular deviation from 

354 LIGHT. 

the object is equal to twice the inclination of the mirrors. Let 
A BCD (Fig. 308) be two plane mirrors inclined at the angle 
AGC. If an eye at H sees the 
star Sin the direction 0, the an- s/ ] 

^ABS= GBD-, :.HBD = 
2GBD. In like manner, B D 

-HBD = ZBDC - 2 GBD 

This principle is employed in 
the construction of Hartley s quad 
rant, and the sextant, used at sea 
for measuring angular distances. 
The angles measured are twice as 
great as the arc passed over by 
the index which carries the re 
volving mirror ; hence, in the 

quadrant, an arc of 45 is graduated into 90 ; and, in the sextant, 
an arc of 60 is graduated into 120. 

596. Multiplied Images by Two Mirrors. 

1. Parallel Mirrors. The series of images is infinite in num 
ber, and arranged in a straight line, perpendicular to the mirrors. 
The object E between the parallel mirrors, AB, CD (Fig. 309), 

FIG. 309. 

J* & f\JS & ? r $* 

-V- ........ V-T-V ....... V- 

M J 

has an image at F, as far behind A B as E is in front of it, and 
between the same perpendiculars. The rays reflected by A B 
diverge, as though they emanated from F; hence, F may be re 
garded as an object before CD, whose image is at F , as far behind 
it. Again, F may be considered as an object before A B, and so 
on indefinitely. Another series exists in the same line, by begin 
ning with G, the first image behind CD. As light is absorbed 
and scattered by each reflection, these images grow fainter, and at 
length disappear. Articles of jewelry are sometimes apparently 
multiplied and extended over a large surface, by lining the cases 
with parallel mirrors. 


The multiplied images of a small bright object, sometimes 
seen in a looking-glass, are produced by repeated reflections be 
tween the front and the silvered covering on the back side. At 
each internal impact on the first surface some light escapes, and 
shows us an image, while another portion is reflected to the back, 
and thence forward again. The image of a lamp viewed very 
obliquely in a mirror is sometimes repeated eight or ten times ; 
and a planet, or bright star, when seen in a looking-glass, will 
be accompanied by three or four faint images, caused in the 
same way. 

2. Inclined Mirrors. In this case, the images are limited in 
number, and arranged in the circumference of a circle, whose 
centre is in the line of common section of the planes of the mir 
rors, and whose radius is the distance of the object from that line. 
Let A B, A C (Fig, 310) be the mirrors, and E the object. Draw 
E G perpendicular to A B, 
and make E F F G, then 
will G be the first image : in 
the same way, find /, the im 
age of G by A C\ /i, the 
image of /; and V, that of 
K. Then begin with the 
mirror A C, and find, as be 
fore, M, 0, P, Q, the succes 
sive images by the two mir 
rors. Ko image of V or Q 
can be formed, because they 
are tfehind both mirrors. All 
these images are in the cir 
cumference of a circle, whose 
radius is E A ; for E F, F A, 
and angle at F, are respectively equal to G F, F A, and angle at 
F\ :. E A = G A ; and in the same way it may be proved, that 
E A A M, A /, &c. If the edges at A be separated, making 
the inclination of the mirrors less and less, the number of images 
will increase, and the circumference will approach a straight line, 
so that ultimately we shall have the case described in (1), in which 
the mirrors are parallel. 

597. Path of the Pencil by which each Image is Seen. 

Fig. 311 will assist to understand how each image is seen by a 
pencil of light which passes back and forth between the mirrors, 
until it reaches the eye. If the eye is at 0, and the object at Q, 
and its images at A, B, (7, D, each image is of course seen by a 
pencil which comes from the mirror to the eye, as if it originated 



FIG. 311. 

in that image. Therefore, draw a line from any image as D, to 
the eye, and from its intersection with the mirror draw a line 
to the preceding image ; 
from the intersection of 
that line with the other 
mirror, a line to the im 
age next preceding, and 
so on back to (); the 
whole path of the pen 
cil will then be traced. 
Thus, A being joined, 
and Q a drawn to the in 
tersection a, the image A 
is seen by the ray Q a, 
a 0. In like manner, B is 
seen by Q , b c, c ; (7, 
by Q d, d e, ef,fO; and 
A by Q g, g h, h i, i Jc, 

598. The Kaleidoscope. This instrument, when carefully 
constructed, beautifully exhibits the phenomenon of multiplied 
reflection by inclined mirrors. It consists of a tube containing 
two long, narrow, metallic mirrors, inclined at a suitable angle ; 
and is used by placing the objects (fragments of colored glass, 
etc.) at one end, and apptying the eye to the other. In order that 
there may be perfect symmetry in the figure made up of the ob 
jects and their successive images, the angle of the mirrors should 
be of such size, that it can be exactly contained an even number 
of times in 360 Q . The best inclination is 30 ; and the field of 
view is then composed of 12 sectors. It is also essential, that the 
small objects forming the picture, should lie at the least possible 
distance beyond the mirrors. To insert three mirrors instead of 
two, as is often done, only serves to confuse the picture, and mar 
its beauty. 

599. Images by the Concave Mirror. The concave mir 
ror forms various images, either real or apparent, either greater or 
less than the object, either erect or inverted, according to the place 
of the object. 

1. The object between the mirror and its principal focus. By 
Art. 588 (2), rays which diverge from a point between the mirror 
and its principal focus, continue to diverge after reflection, but in 
a less degree. Let C be the centre, and F the principal focus of 
the mirror M N (Fig. 312), and A B the object. Draw the axes, 



C A, C B, and produce them behind the mirror. The pencil from 
A will be reflected to the eye at //, radiating as from a, in the 
same axis; likewise, those from B, as from b. Therefore, the 

FIG. 312. 


image is apparent, since rays do not actually flow from it ; erect, 
as the axes do not cross each other between the object and image ; 
enlarged, because it subtends the angle of the axes at a greater dis 
tance than the object does. As the object approaches, and finally 
reaches the principal focus, the reflected rays approach parallelism, 
and the image departs from the mirror, till it is at an infinite dis 
tance, and is viewed as a heavenly body. 

2. Object between the principal focus and the centre. As soon 
as the object passes the principal focus, the rays of each pencil be 
gin to converge ; and each radiant of the object has its conjugate 
focus in the same axis beyond the centre (Art. 589). For exam 
ple, the pencil A dg (Fig. 313) is converged to a in the axis A Ca, 
and B D G to b, in the axis B Cb. Therefore, the image of A B 
is a b beyond the centre ; and if an observer is beyond a b, the 
rays, after crossing at the image, will reach him, as though they 

FIG. 313. 

originated in a b ; or if a screen is placed at a b, the light which 
is collected in the focal points will be thrown in all directions by 
radiant reflection from the screen. Hence, the image is real ; it 
is also inverted, because the axes cross between the conjugate foci ; 
and it is enlarged, since it subtends the angle of the axes at a 
greater distance than the object does. That b C is greater than 
B C, is proved by joining C G, which bisects the angle B G b, ami 
therefore divides B b so that B C : Cb :: B G : Gb. When the 
object reaches the centre, the image is there also, but inverted in 
position, since rays which proceed from one side of C, are reflected 
to the other side of it. 

358 LIGHT. 

3. Object beyond the centre. This is the reverse of (2), the 
conjugate foci having changed places ; a b, therefore, being the 
object, A B is its image, real, inverted, diminished. As the ob 
ject removes to infinity, the image proceeds only to the principal 
focus F. 

600. Illustrated by Experiment These cases are shown 
experimentally by placing a lamp close to the mirror, and then 
carrying it along the axis to a considerable distance away. While 
the lamp moves from the mirror to the principal focus, its image 
behind the mirror recedes from its surface to infinity ; we may 
then regard it as being either at an infinite distance behind, or an 
infinite distance in front, since the rays of every pencil are par 
allel. After the lamp passes the principal focus, the image ap 
pears in the air at a great distance in front, and of great size, and 
they both reach the centre together, where they pass each other ; 
and, as the lamp is carried to great distances, the image, growing 
less and less, approaches the principal focus, and is there reduced 
to its smallest size. The only part of the infinite line of the axis 
before and behind, in which no image can appear, is the small dis 
tance between the mirror and its principal focus. 

If a person looks at himself, so long as he is between the mir 
ror and the principal focus, he sees his image behind the mirror 
and enlarged. But when he is between the principal focus and 
centre, the image is real, and behind him ; the converging rays of 
the pencils, however, enter his eyes, and give an indistinct view of 
his image as if at the mirror. When he reaches the centre, the 
pupil of the eye is seen covering the entire mirror, because rays 
from the centre are perpendicular, and return to it from all parts 
of the surface. Beyond the centre, he sees the real image in the 
air before him, distinct and inverted. 

601. Images by the Convex Mirror. The convex mirror 
affords no variety of cases, because diverging rays, which fall upon 

Fia. 314. 

it, are made to diverge still more by reflection. - In Fig. 314, the 
pencil from A is reflected, as if radiating from a in the same axis 


A C, and that from 7>, as from # in the axis B C, and these ap 
parent radiants are always nearer the surface than the middle 
point between it and C (Art. 590). The image is therefore appar 
ent ; it is erect, since the axes do not cross between the object and 
image ; and it is diminished, as it subtends the angle of the axes 
at a less distance than the object 

602. Caustics by Reflection. These are luminous curved 
surfaces, formed by the intersections of rays reflected from a hemi 
spherical concave mirror. The. name caustic is given from the 
circumstance that heat, as well as light, is concentrated in the 
focal points which compose it. BAD 

(Fig. 315), represents a section of the FlG - 

mirror, and B F D of the caustic ; the 
point F, where all the sections of the 
caustic through the axis meet each 
other, is called the cusp. When the 
incident rays are parallel, as in the 
iigure, the cusp is at the principal 
focus, that is, the middle point be 
tween A and C. The rays near the 
axis R A, after reflection meet at the 
cusp (Art. 588) ; but those a little more 
distant cross them, and meet the axis a little further toward A. 
And the more distant the incident ray from the axis, the further 
from the centre does the reflected ray meet the axis. Thus each 
ray intersects all the previous ones, and this series of intersections 
constitutes the curve, B F. The curve is luminous, because it 
consists of the foci of the successive pencils reflected from the 
arc A B. 

If the incident rays, instead of being parallel, diverge from a 
lamp near by, the form of the caustic is a little altered, and the 
cusp is nearer the centre. This case may be seen on the surface 
of milk, the light of the lamp being reflected by the edge of the 
bowl which contains it 

If parallel or divergent light falls on a convex hemispherical 
mirror, there will be apparent caustics behind the mirror ; that is, 
the light will be reflected as if it radiated from points arranged in 
such curves. 

603. Spherical Aberration of Mirrors. It has already 
]>een mentioned (Art. 587), that the statements in this chapter 
relating to focal points and images, as produced by spherical mir 
rors, are true only when the mirror is a very small part of the 
whole spherical surface. In Art G02 we have seen the effect of 

360 LIGHT. 

using a large part of the spherical surface viz., the rays neither 
converge to, nor diverge from a single point, but a series of points 
arranged in a curve. This general effect is called the spherical 
aberration of a mirror ; since the deviation of the rays is due to 
the spherical curvature. The deviation, as we have seen, is quite 
apparent in a hemisphere, or any considerable portion of one ; but 
it exists in some degree in any spherical mirror, unless infinitely 
small compared with the hemisphere. 

But there are curves which will reflect without aberration. 
Let a concave mirror be ground to the form of a paraboloid, and 
rays parallel to its axis will be converged to the focus without 
aberration. For, at any point on such a mirror, a line parallel to 
the axis, and a line drawn to the focus, make equal angles with 
the tangent, and therefore, equal angles with the perpendicular to 
the surface. And rays, parallel to the axis of a convex paraboloid, 
will diverge as if from its focus, on the same account. Again, if a 
radiant is placed at the focus of a concave parabolic mirror, the 
reflected rays will be parallel to the axis, and will illuminate at a 
great distance in that direction. Such a mirror, with a lamp in 
its focus, is placed in front of the locomotive engine to light the 
track, and has been much used in light-houses. If a concave mir 
ror is ellipsoidal, light emanating from one focus is collected with 
out aberration to the other, because lines from the foci to any point 
of the curve make equal angles with the tangent at that point. 

Since heat is reflected according to the same law as light, a 
concave mirror is a burning-glass. When it faces the sun, the 
light and heat are both collected in a small image of the sun at 
the principal focus. And, if no heat were lost by the reflection, 
the intensity at the focus would be to that of the direct rays, as 
the area of the mirror to the area of the sun s image. Burning 
mirrors have sometimes been constructed on a large scale, by giv 
ing a concave arrangement to a great number of plane mirrors. 



604. Division of the Incident Beam. When light falls 
on an opaque body, we have noticed that it is arrested, and a 
shadow formed beyond. Of the light thus arrested, a portion is 
reflected, and another portion lost, which is said to be absorbed by 


the body. When light meets a transparent body, a part is still 
reflected and a small portion absorbed, but, in general, the greater 
part is transmitted. The ratio of intensities in the reflected and 
transmitted Beams varies with the angle of incidence, but little 
being reflected at small angles of incidence, and almost the whole 
at angles near 90. 

605. Refraction. The transmitted beam suffers important 
changes, one of which is a change in direction. This change is 
called refraction, and takes place at the surface of a new medium. 
In Fig. 316, A C, incident upon R S, the surface of a different 
medium, is turned at C into another 
line, as C E, which is called the re- FIG. 316. 

fracted ray. The angle E. C Q, be 
tween the refracted ray and the perpen 
dicular is called the angle of refraction ; 
the angle G C E, between the direc 
tions of the incident and the refracted 
rays, is the angle of deviation. 

It is a general fact, to which there 
are but few exceptions, that a ray of 

light in passing out of a rarer into a denser medium is refracted 
toward the perpendicular to the surface ; and in passing out of a 
denser into a rarer medium, it is refracted from the perpendicular. 
But the chemical constitution of bodies sometimes affects their 
refracting power. Some inflammable bodies, as sulphur, amber, 
and certain oils, have a great refracting power in comparison with 
other bodies ; and in a given instance, a ray of light in passing 
out of one of these substances into another of greater density may 
be turned from the perpendicular instead of toward it. In the 
optical use of the words, therefore, denser is understood to mean, 
of greater refractive power ; and rarer signifies, of less refractive 
poiver. In Fig. 316, the medium below R S is of greater refrac 
tive power than that above. 

We see an example of refraction in the bent appearance of an 
oar in the water, the light which comes to the eye from the part 
immersed is bent from the perpendicular as it passes from water 
into air, and causes it to appear higher than its true place. In the 
same manner, the bottom of a river appears elevated, and dimin 
ishes th % e apparent depth of the stream. Let a small object be 
placed in the bottom of a bowl, and let the eye be withdrawn till 
the object is hidden from view by the edge of the bowl. If now 
the bowl be filled up with water, the object is no longer concealed, 
for the light, as it emerges from the water, is bent away from the 
perpendicular, and brought low enough to enter the eye. 



696. Law of Refraction. The law which is found to hold 
true in all cases of common refraction is this : 

The angles of incidence and refraction are on opposite sides of 
the perpendicular to the surface, and, for any given inertia, the sines 
of the angles have a constant ratio for all inclinations. 

For example, in Fig. 317, if A C is refracted to E, then a G 
will be refracted to e, so that AD-.EF 
.-. a d: ef\ and if the rays pass out in 
a contrary direction, the ratio is also 
constant, being the reciprocal of the 
former, viz., E F : A D : : e f : a d. 

A ray perpendicular to the surface, 
passing in either direction, is not re 
fracted; for, according to the law, if 
the sine of one angle is zero, the sine of 
the other must be zero also. Which 
ever way light passes, when air is one 
of the media, suppose the sine of the 
smaller angle, i. e. the angle in the denser medium, to be 1, then 
the sine of the larger angle for water is 1.336 ; and for crown glass, 
it is about 1.5. The number, in each case, expresses the constant 
ratio of the sines, for the given media, and is called the index of 
refraction, and is employed as the measure of refractive power. 
The following table gives the index of refraction for a few sub 
stances : 

Chromate of lead, . . . 2.974 
Red silver ore, . . . .2.564 

Diamond, 2.439 

Phosphorus, 2.224 

Sulphur, ...... 2.148 

Flint glass, 1-830 

Sapphire, 1.800 

Sulphuret of carbon, . .1.768 

Oil of cassia, 1.641 

Quartz, 1-548 

Amber, !-547 

Crown glass, 1-53 

Oil of olives, 1470 

Alum, x -457 

Fluorspar, 1-434 

Mineral acids, . . . .1.410 

Alcohol, 1.372 

Water, 1.336 

Ice, 1.309 

Tabasheer, i.m 

607. Limit of Transmission from 
a Denser to a Rarer Medium. As a 

consequence of the law of refraction, 
there is a limit beyond which a ray can 
not escape from a denser medium. Let 
A C (Fig. 318) be the ray incident upon 
the rarer medium R E S. It will be 
refracted from the perpendicular D F 
into the direction C E, so that A D is to 

FIG. 318. 


E F in a constant ratio (Art. 606). If the angle A C D be 
increased, the angle F C E must also increase till its sine equals 
C S. Make ad\ C S\\AD\ E F\ then a C is the limit of the inci 
dent rays which can emerge. For if a C D is enlarged, its sine 
is increased, and therefore the sine of refraction must increase ; 
but this is impossible, since it is already equal to the radius C S. 
Hence it follows, that whenever the angle of incidence is greater 
than that at which the sine of the angle of refraction becomes 
equal to radius, the ray cannot be refracted consistently with the 
constant ratio of the sines. 

This is proved also by experiment ; the emerging ray increases 
its angle of refraction till it at length ceases to pass out. Beyond 
that limit all the incident rays are reflected from the inner surface 
of the denser medium ; and this reflection is more perfect than 
any external reflection, and is called total reflection. If n = the 
index of refraction, the limit at which refraction ceases and total 
reflection begins is found by the proportion, n : 1 : : rad. : sine of 
the limit. If the refractive power is greater, the limit is smaller ; 
for, by the above proportion, since the means are constant, n varies 
inversely as sine of limit. For water, it is 48 28 ; for crown glass, 
40 49 ; for diamond, 24 12 . 

608. Transmission through Plane Surfaces. 

1. A medium bounded by parallel planes. In this case the 
incident and emergent rays are parallel. Let D E (Fig 319) enter 
the medium A B b a at E, and leave 

it at F, and let P Q, R 8 be the FlG - 319 - 

perpendiculars at E, F.- The first 
angle of refraction Q E F, and the 
second angle of incidence, E F R, 
are equal, being alternate; there 
fore, DEP=SFG, since their 
sines have a constant ratio to those 
of Q E F, E F R. Hence, if the 
incident rays are produced to C and 
H, the angles of deviation are equal ; but D E Fis supplement to 
the first angle of deviation, and FlG 320. 

E F 6; of the second. There 
fore, as D E .Fand EFG are 
alternate and equal, D E is 
parallel to F G. 

2. A medium bounded by 
inclined planes, called a prism. 
The transmitted ray is turned 
from the refracting angle. Let 

364 LIGHT. 

ABC (Fig. 320) be that section of a glass prism which is per 
pendicular to its axis, and A C, B O, the inclined sides of it, 
through which the light is transmitted. is called the refracting 
angle, and A B the base. As prisms are usually constructed and 
mounted, either A, B, or C may be the refracting angle ; but it is 
not essential that any of the faces should meet at an edge, as the 
effect on the light depends only on the inclination. In ordinary 
directions of the ray, the two refractions, one on entering, the 
other on leaving the prism, conspire to increase the deviation of 
the ray from its original direction. D E is first bent toward E K, 
making the deviation HE F\ at F, it is turned from F Q, making 
a second deviation, E F 1, the same way. The sum of the two 
deviations, IEF+EFI= Gf I II, the total deviation away 
from the refracting angle, C. 

To an eye at G, the radiant D is seen in the direction G F I. 

609. The Multiplying Glass. A piece of glass ground 
with one side plane, and the other in any number of plane facets 
on a convex surface, is called a multiplying glass Each facet, 
along with the opposite plane surface, forms a prism; and if a 
radiant A is placed in the axis (a perpendicular through the cen 
tre of the plane surface), the 

pencils, falling on the several FIG. 321. 

facets, will be turned from 

the edge, and may by two 

refractions at the opposite 

surfaces be brought to an eye 

placed also in the axis, and 

thus as many images will be 

seen as there are facets. Fig. 

321 exhibits the effect of 

seven such facets. 

610. Prism used for Measuring Refractive Power. 

The following theorem may be used for determining the refractive 
power of a substance, after first forming it into a prism of small 
angle : 

If the angle of deviation be divided ~by the refracting angle of the 
prism, and the quotient le added to unity, the sum is the index of 

In proving this, it is assumed that all the angles are very 
small, so that they vary as their sines. Let n = the index of re 
fraction, then (Fig. 320), 

E E I(=D E P}\KE F\ \n\\\ .-. F E 1\ K E F \\n-\\\\ 
also KFI(=GFQ)\ KFE\ \n\\\ :. EFI\ KFE\ : n-1 ; 1 ; 


FEI+EFItEEF+EFEnn-l .!; 
. . FIHiP KF;:n-l -I. 


But P K F A C B, each being the supplement of E K F. 
Therefore, Fl II : A C B : : n - I : 1 ; 


~ACB> "ACS* 


Now, in crown glass, . , .-. is found by trial to be yery near J- ; 

/. n = 1.5 nearly. 

In order to find the index of refraction for any solid substance, 
grind it into a prism whose sides are nearly parallel, and carefully 
measure their inclination . Then measure the displacement cf a 
distant object seen through it at right angles to its surface. For 
example, the faces of a transparent mineral incline 1 10 ; and 
when held before the eye, it displaces a distant object 50 ; .-. the 
index of refraction = 1 + f = 1.714. 

611. Light through one Surface. 

1. Plane Surface. When parallel rays pass into another me 
dium through a plane surface, they remain parallel. For the per 
pendiculars being parallel, the angles of incidence are equal, and 
bherefore the angles of refraction are equal also, and the refracted 
rays parallel. But a pencil of diverging rays is made to diverge 
less, when it enters a denser medium. For the outer rays make 
the largest angles of incidence, and are therefore most refracted 
toward the perpendiculars, and thus toward parallelism with each 
other. And when diverging rays enter a rarer medium, they di 
verge more ; because the outside rays make the largest angles of 
incidence, and therefore the largest angles of refrac 
tion, by which means they spread more from each 

The last case is illustrated when we look per 
pendicularly into water, and see its depth apparently 
diminished by about one-fourth of the whole. Let 
A B (Fig. 322) be the surface, and C a point at the 
bottom, from which a pencil comes to the eye. Let 
C F, the axis of the pencil, be perpendicular to A B, 
and C B E an oblique ray of the pencil. The an 
gle C C B H angle of incidence ; and A D B 
GEE angle of refraction. Now, in the tri 
angle E D C, B C\ B D (: A C: A D nearly) . : 
sm D : sin C : : sine of refraction ; sine of incidence 
: . 1.34 : 1. Hence the apparent depth is one-fourth 
less than the real depth. The apparent depth of 



water may be diminished much more than this by looking into it 

2. Convex surface of the denser. A convex surface tends to 
converge rays. Let C (Fig. 323) be the centre of convexity, and 
C D, C C } two radii produced. As rays are bent toward the per. 

FIG. 323. 

pendiculars in entering a denser medium, and as the perpendicu 
lars themselves converge to C , the general effect of such a surface 
is to produce convergency. The pencil, A H, A N, is merely 
made less divergent, H D , N A ; B ff, B N become parallel, 
H D , N B ; D H, D N, convergent to D ; the parallel rays, D H, 
EN, convergent to E { ; the convergent pencil, D II, F N, more 
convergent to F ; but D H, C N, which converge equally with 
the radii, are not changed ; and D H, G N, which converge more 
than the radii, converge less than before, to G . The two last 
cases, which are exceptions to the general effect, rarely occur in 
the practical use of lenses. 

If we trace in the opposite direction the rays, A , B , D r , &c., 
comparing each with D D, we find, in this case also, that the 
convex surface tends to converge the rays, by bending them from 
C D, C C. 

3. Concave surface of tlie denser. A concave surface tends to 
diverge rays. Let C C , CD (Fig. 324), be the radii of concavity 
produced. As the radii diverge in the direction in which the light 

FIG. 324. 

moves, the rays, being bent toward them, will generally be made 
to diverge also. Hence, parallel rays, B H, E N, are diverged, 
H D, N E ; and diverging rays, B II, B N, are diverged more, 
H D, N B . If, however, rays diverge as much as the radii, or 



more, they proceed in the same direction, or diverge less, a case 
which rarely occurs. 

If the rays are traced in the opposite direction, the tendency 
in general to produce divergency appears from the fact that the 
perpendiculars are now converging lines, and the rays arc refracted 
from them. 

612. Lenses. A lens is a circular piece of glass, whose sur 
faces are plane or spherical, and the spherical surface either convex 
or concave. The usual varieties are shown in Fig. 325. 

FIG. 325. 

A doulle convex lens (A) consists of two spherical segments, 
either equally or unequally convex, having a common base. 

A plano-convex lens (B) is a lens having one of its sides con 
vex and the other plane, being simply a segment of a sphere. 

A double concave lens (C) is a solid bounded by two concave 
spherical surfaces, which may be either equally or unequally 

A plano-concave lens (D) is a lens one of whose surfaces is 
plane and the other concave. 

A meniscus (E] is a lens one of whose surfaces is convex and 
the other concave, but the concavity being less than the convexity, 
it takes the form of a crescent, and has the effect of a convex lens 
whose convexity is equal to the difference between the sphericities 
of the two sides. 

A concavo-convex lens (F) is a lens one of whose surfaces is 
convex and the other concave, the concavity exceeding the con 
vexity, and the lens being therefore equivalent to a concave lens 
whose concavity is equal to the difference between the sphericities 
of the two sides. 

A line (3f N) passing through a lens, perpendicular to its op 
posite surfaces, is called the axis. The axis usually, though not 
necessarily, passes through the centre of the figure. 

613. General Effect of the Convex Lens. Whether dou 
ble-convex or plano-convex, its general effect is to converge light. 
It has been shown (Art. Gil) that the convex surface of a denser 



medium tends to converge rays, whichever way they pass through 
it. Therefore, if E (Fig. 326) is a radiant, while E C C follows 

FIG. 326. 

the axis without change of direction, the oblique ray E D is first 
refracted toward D C, and then from C D produced, and both 
actions conspire to converge it to the axis. The rays are repre 
sented as meeting in the focus F. "Whether the rays are actually 
converged, depends on their previous relation to each other. If 
the lens is plano-convex, the plane surface has usually but little 
effect in converging the light ; but by Art. 611 it may be shown 
that its action will usually conspire with that of the convex 

614. General Effect of the Concave Lens. This lens, 
whether double-concave or plano-concave, tends to produce diver 
gency. This is evident from what has been shown in Art. 611. 
The ray ED (Fig. 327), in entering the denser medium, is first 

FIG. 327. 

refracted toivard C D produced, and on leaving the medium at /) , 
is refracted from D C ; and is thus twice refracted from the ray 
E C, which being in the axis, is not refracted at all. If the lens 
is plano-concave, the effect of the plane surface may, or may not, 
conspire with that of the concave surface. 

615. The Optic Centre of a Lens. Within every lens 
there is a point called the optic centre, so situated that the inci 
dent and emergent portions of every ray which passes through it 
are parallel to each other. Let C, C (Fig. 328), be the centres of 
the two surfaces of the lens ; draw the axis C C , also any oblique 



radius C A, and C B parallel to it ; then join A B ; the point E, 
in which A B intersects the axis, is the optic centre, and R A the 
incident, and B R the emer 
gent portion of the ray passing FlG - 
through A and B, are parallel 

to each other. For the angles *> M ^4fiZ ; ^ C 

UAC, and E B C , are equal, 
being alternate, and therefore 
the ray is refracted at A and B 

equally and in opposite directions, making R A and B R par 
allel, as proved in Art. 608, 1. But the point E is the same, 
whatever may be the points A and B, to which the parallel radii, 
C A, C B, are drawn. For, since the triangles, EAC,EBC , 
are similar, GA : C B : : C E : C 1 E; . . CA + C B : C B : : CE 
+ C E : C" E ; and as the three first terms are constant, the 
fourth, (7 E, is constant also, and E is a fixed point. 

When the lens is thin, and the rays are nearly parallel to its 
axis, the ray R A B R may be considered a straight line ; and it 
forms the axis of the pencil of light which passes through the lens 
in that direction. 

616. Conjugate Foci. If the rays from R (Fig. 329) are 
collected at F, then rays emanating from F will be returned 
to R\ and the two points are called conjugate foci. Their rela 
tive distances from the lens may be determined when the radii of 

the surfaces and the index of refraction are known. Let n be the 
index of refraction, and assume, what is practically true, that the 
angles of incidence and refraction are so small that their ratio is 
the same as the ratio of their sines. Then 


n - 1 : 1. 


.-. KGH\IGH:\n- 1 

in like manner K H G : I H G :\n 1 

.-. KGff+ KH G .I GH + IHG: 

and IGH+IIIG = GIC= C + 

370 LIGHT. 

naming the acute angles at R, C, C } F, by those letters re 

.-. R + F:C + C ::n- 1:1. 

Now, the lens being thin, and the angles R, C, C , and F very 
small, the same perpendicular to the axis, at L, the centre of the 
lens, may be considered as subtending all those angles. Hence, 
each angle is as the reciprocal of its distance from L. Let R L 
p; F L = q; C L = r\ and C L r . Then the equation above 

1 + 1:M,:: l:l; 

p q r r 

which expresses in general the relation of the conjugate foci. To 
adapt it to crown-glass, call n = f , and we have 


p q r r 

617. To find the Principal Focus. The radiant from 
which parallel rays come is at an infinite distance. Therefore, 
making p = oc , and the distance of the principal focus = F, we 

have - = 0, and 

f r crown ass this is 

(n - 1) (r + r ) 

If the curvatures are equal, F . - -^- ; for crown-glass this 

A (n Ij 

becomes F = r ; that is, the principal focus of a double convex 
lens of crown-glass, having equal curvatures, is at the centre of 

The foregoing formulae are readily adapted to the other forms 
of lens. When a surface is plane, its radius is infinite, and 

-, or = 0. When concave, its centre is thrown upon the same 

side as the surface, and its radius is to be called negative. And if 
the focal distance, as given by the formula, becomes negative, it is 
understood to be on the same side as the radiant ; that is, the 
focus is a virtual radiant. 

618. Images by the Convex Lens. The convex lens forms 
a variety of images, whose character and position depend on the 
place of the object. If it is at the principal focus, the rays of every 



pencil pass out parallel, and seem to come from an infinite dis 
tance. If the object is nearer than the principal focus, the emer 
gent rays of each pencil diverge less than the incident rays, and 
therefore seem to radiate from points further back ; the image is 
therefore apparent. Let M N (Fig. 330) be the object nearer than 
the principal focus, F. Then the pencil from M will, after refrac- 

FIQ. 330. 

tion, diverge as from mm C M produced, and so of every point ; 
hence m n is the image. It is erect, because the axes of the pencils 
do not cross between the object and image ; and it is enlarged, 
because it subtends the angle M C N at a greater distance than 
the object. 

But if the object is further from the lens than the principal 
focus, the rays of each pencil converge to a point in the axis of 
that pencil produced through the lens ; and thus light is collected 
in focal points, which consequently become actual radiants. The 
last case is illustrated by Fig. 331, in which M Nis the object, and 

FIG. 331. 

m n the image. A cone of rays from covers the lens L L, and 
is converged again into the axis at 0, the conjugate focus of 0, and 
there cross, and proceed as from a radiant. The cone of rays from 
M is converged to m in the axis Mm of that cone, which is a 
straight line through the optic centre (Art. 615) ; and so from 
every point of the object. Though the rays of every radiant con 
verge from the lens to the conjugate focus of that radiant, yet the 
axes of the pencils diverge from each other, having all crossed at 
the optic centre. The image is therefore inverted, as are all real 
images, in whatever way produced. 



The formula for conjugate foci shows that if p is increased, q 
is diminished ; therefore the further M N is removed from the 
lens, the nearer m n approaches to it ; but the nearest position is 
the principal focus, which it reaches when the object is at an infi 
nite distance. As the object and image subtend equal angles at 
the optic centre, and are parallel, or nearly parallel with each 
other, their diameters are proportioned to their distances from the 
lens. But the area of the lens has no effect on the size of the 
image, since change of area does not alter the relation of the axes, 
but only the size of the luminous cones, and thus the quantity of 
light in each pencil. 

619. Images by the Concave Lens. As the rays of each 
pencil are diverged more after passing through the lens than before, 
the image is apparent, and is situated between the lens and the 
object. Let M N (Fig. 332) be the object; the cone of rays from 
N will, after refraction, diverge more, as from n, in the same axis 

FIG. 332. 

C N; and all other pencils will be affected in a similar manner, 
and form an apparent image m n. It will be erect, since the axes 
do not cross between, and diminished, being nearer the angle 0, 
which is subtended by both object and image. 

It is noticeable that the concave mirror and the convex lens are 
analogous in their effects, forming images 011 both sides, both real 
and apparent, both erect and inverted, both larger and smaller 
than the object ; while the convex mirror and the concave lens also 
resemble each other, producing images always on one side, always 
apparent, always erect, always smaller than the object. 

620. Caustics by Refraction. If the convex surface of a 
lens is a considerable part of a hemisphere, 
the rays more distant from the axis will be 
so much more refracted than others, as to 
cross them and meet the axis at nearer 
points, thus forming caustics by refraction. 
Fig. 333 shows this effect in the case of 
parallel rays; those near the axis inter- 

FIG. 333. 


seating it at the principal focus F, and the intersections of re 
moter rays being nearer and nearer to the lens, so that the whole 
converging pencil assumes a form resembling a cone with concave 

621. Spherical Aberration of a Lens. The production 
of caustics is an extreme case of what is called spherical aberra 
tion. Unless the lens is of small angular breadth, a pencil whose 
rays originated in one point of an object is not converged accu 
rately to one point of the image, but the outer rays are refracted 
too much, and make their focus nearer the lens than that of the 
central rays, as represented in Fig. 334. If F is the focus of the 
central rays, and F of the 

extreme ones, other rays of FlG< 334> 

the same beam are collected 
in intermediate points, and 
F F is called the longitu 
dinal spherical aberration; 
and G H, the breadth cov 
ered by the pencil at the 
focus of central rays, is called the lateral spherical aberration. 

Such a lens cannot form a distinct image of any object ; be 
cause perfect distinctness requires that all rays from any one point 
of the object should be collected to one point in the image. If, 
for example, the beam whose outside rays are R A, R B, comes 
from a point of the moon s disk, that point will not be perfectly 
represented by F, because a part of its light covers the circle, 
whose diameter is G H, thus overlapping the space representing 
adjacent points of the moon. And if that point had been on the 
edge of the moon s disk, F could not be a point of a well-defined 
edge of the image, since a part of the light would be spread over 
the distance F G outside of it, and destroy the distinctness of its 

622. Remedy for Spherical Aberration. As spherical 
lenses refract too much those rays which pass through the outer 
parts, it is obvious that, to destroy aberration, a lens is required 
whose curvature diminishes toward the edges. Accordingly, forms 
for ellipsoidal lenses have been calculated, which in theory will 
completely remove this species of aberration. But no curved 
solids can be so accurately ground as those whose curvature is uni 
form in all planes, that is, the spherical. Hence, in practice it is 
found better to reduce the aberration as much as possible by spher 
ical lenses, than to attempt an entire removal of it by other forms 
which cannot be well made. 

374 LIGHT. 

In a plano-convex lens, whose plane surface is toward the ob 
ject, the spherical aberration is 4.5; that is (Fig. 334), F F = 4.5 
times the thickness of the lens. But the same lens, with its con 
vex side toward the object, is far better, its aberration being only 
1.17. In a double convex lens of equal curvatures, the aberration 
is 1.67; if the radii of curvature are as 1 : 6, and the most convex 
side is toward the object, the aberration is only 1.07. By placing 
two plano-convex lenses near each other, the aberration may be 
still more reduced. 

623. Atmospheric Refraction. The atmosphere may be 
regarded as a transparent spherical shell, whose density increases 
from its upper surface to the earth. The radii of the earth pro 
duced are the perpendiculars of all the laminae of the air ; and 
rays of light coming from the vacuum beyond, if oblique, are bent 
gradually toward these perpendiculars; and therefore heavenly 
bodies appear more elevated than they really are. The greatest 
elevation by refraction takes place at the horizon, where it is 
about half a degree. 

624. Mirage. This phenomenon, called also looming, con 
sists of the formation of one or more images of a distant object, 
caused by horizontal strata of air of very different densities. Ships 
at sea are sometimes seen when beyond the horizon, and their 
images occasionally assume distorted forms, contracted or elon 
gated in a vertical direction. These effects are generally ascribed 
to extraordinary refraction in horizontal strata, whose difference 
of density is unusually great. But many cases of mirage seem to 
be instances of total reflection from a highly rarefied stratum rest 
ing on the earth. These occur frequently on extended sandy 
plains, as those of Egypt. When the surface becomes heated, 
distant villages, on more elevated ground, are seen accompanied 
by their images inverted below them, as in water. As the traveler 
advances, what appeared to be an expanse of water retires before 
him. By placing alcohol upon water in a glass vessel, and allow 
ing them time to mingle a little at their common surface, the phe 
nomena of mirage may be artificially represented. 





625. The Prismatic Spectrum. Another change which 
light suffers in passing into a new medium, is called decomposi 
tion, or the separation of light into colors. For this purpose, the 
glass prism is generally employed. It is so mounted on a jointed 
stand, that it can be placed in any desired position across the 
beam from the heliostat The beam, as already noticed, is bent 
away from the refracting angle, both in entering and leaving the 
prism, and deviates several degrees from its former direction. If 
the light is admitted through a narrow aperture, F (Fig. 335), and 

FIG. 335. 

the axis of the prism is placed parallel to the length of the aper 
ture, the light no longer falls, as before, in a narrow line, L, but 
is extended into a band of colors, R V, whose length is in a plane 
at right angles to the axis of the prism. This is called the pris 
matic spectrum. Its colors are usually regarded as seven in 
number red, orange, yellow, green, Hue, indigo, violet. The red 
is invariably nearest to the original direction of the beam, and 
the violet the most remote ; and it is because the elements of 
white light are unequally refrangible, that they become separated, 
by transmission through a refracting body. The spectrum is 
properly regarded as consisting of innumerable shades of color. 
Instead of Newton s division into seven colors, many choose to 
consider all the varieties of tint as caused by the combination of 
three primitive colors, red, yellow, and blue, varying in their pro- 



portions throughout the entire spectrum. The number seven, 
as perhaps any other particular number, must be regarded as 

The spectrum contains other elements besides those which 
affect the eye. If a thermometer bulb be moved along the spec 
trum, it is found that the greatest heat lies outside of the visible 
spectrum at the red extremity ; hence heat is less refrangible than 
light. On the other hand, the chemical or actinic rays are more 
refrangible than the luminous rays, and fall at and beyond the 
violet end of the spectrum. 

Light from other sources is also susceptible of decomposition 
by the prism ; but the spectrum, though resembling that of the 
sun, usually differs in the proportion of the colors. 

626. The Individual Colors of the Spectrum cannot be 
Decomposed by Refraction. Some of the colors of the spec 
trum are called simple colors, namely, red, yellotu, and blue, while 
the others are generally regarded as compound colors ; for orange 
may be formed of red and yellow, green of yellow and blue, while 
indigo and violet are mixtures of blue and red in different propor 
tions. It is nevertheless true, that none of the colors of the spec 
trum can be decomposed by refraction. For if the spectrum 
formed by the prism A be allowed to fall on the screen ED (Fig. 
336), and one color of it, green for example^ be let through the 

FIG. 336. 


screen, and received on a second prism, J9, it is still refracted as 
before, but all its rays remain together and of the same color. 
The same is true of every color of the spectrum. Therefore, so 
far as refrangibility is concerned, all the colors of the spectrum 
are alike simple. 

627. Colors of the Spectrum Recombined. It may be 

shown, in several ways, that if all the colors of the spectrum be 
combined, they will reproduce white light. One method is by 
transmitting the beam successively through two prisms whose 
refracting angles are on opposite sides. By the first prism, the 
colors are separated at a certain angle of deviation, and then Ml 



on the second, which tends to produce the same 
deviation in the opposite direction, by which 
means all the colors are brought upon the same 
ground, and the illuminated spot is white as if no 
prism had been interposed. Or the colors may 
be received on a series of small plane mirrors, 
which admit of such adjustment as to reflect all 
the beams upon one spot. Or finally, the several 
colors can, by different methods, be passed so rap 
idly before the eye that their visual impressions 
shall be united in one ; in which case the illu 
minated surface appears white. 

628. Complementary Colors. If certain 
colors of the spectrum are combined in a com 
pound color, and the others in another, these two 
are called complementary colors, because, when 
united, they will produce white. For example, if 
green, Uue, and yelloiu are combined, they will 
produce green, differing slightly from that of the 
spectrum ; the remaining colors, red, orange, in 
digo, and violet, compose a kind of purple, unlike 
any color of the spectrum. But these particular 
shades of green and purple, if mingled, will make 
perfectly white light, and are therefore comple 
mentary colors. 

629. Fixed Dark Lines of the Spectrum. 

Let the aperture through which the sunbeam 
enters be made exceedingly narrow, and let the 
prism be of uniform density, and then let the re 
fracted pencil pass immediately through a small 
telescope, and thence into the eye, and there ap 
pears a phenomenon of great interest the dark 
lines, or the Fraunhofer lines, as they are often 
called from the name of their discoverer. These 
lines, an imperfect view of which is presented in 
Fig. 337, are unequal in breadth, in darkness, and 
in distance from each other, and so fine and 
crowded in many parts that the whole number 
cannot be counted. Fraunhofer himself described 
between 500 and 600, among which a few of the 
most prominent are marked by letters, and used 
in measuring refractive power. At least as many 
as six thousand are now known and mapped, so 

FIG. 337. 


378 LIGHT. 

that any one of them may be identified. They are parallel to each 
other, and perpendicular to the length of the spectrum. When 
the pencil passes through a succession of prisms, all bending it 
the same way, the spectrum becomes more dilated, and more lines 
are seen. The instrument fitted up as above described, either with 
one prism or a series of prisms, is called a spectroscope. 

630. Bright Lines in the Spectrum of Flame. If the 

spectroscope be used for the examination of the flame of different 
substances in combustion, the spectrum is found to consist of cer 
tain bright lines, differing in color and number, according to the 
substance under examination. Thus, the spectrum of sodium 
flame, besides showing other fainter lines, consists mainly of two 
conspicuous yelloiv lines, very close together, so as ordinarily to 
appear as one. The flame of carbon shows two distinct lines, one 
of which is green, the other indigo. In this respect every sub 
stance differs from every other, and each may be as readily distin 
guished by the lines which compose its spectrum as by any other 
property. The lines of some substances are very numerous ; as, 
for example, iron, whose spectrum lines amount to four or five 

But a solid or liquid substance, when raised to a red or white 
heat, without passing into the gaseous state and producing flame, 
forms a continuous spectrum, having neither bright nor dark 

631. The Spectrum of a Heated Solid or Liquid Shin 
ing through Flame. The condition of a spectrum is entirely 
changed when the light from a heated solid or liquid substance 
shines through the flame of a burning gas. The bright lines 
instantly become dark lines. The flame seems to absorb just 
those rays, and only those, which are like the rays emitted by 
itself. As an example, the spectrum of sodium flame consists of a 
bright double yellow line, and a few fine luminous lines of other 
colors. Jf now iron at an intense white heat shines through this 
flame, the whole spectrum becomes luminous, except the very 
lines which were before bright ; these are now dark. 

632. Composition of the Sun s Surface. A great number 
of the dark lines of the solar spectrum are identical in position 
with lines in the spectrum of terrestrial substances. The spectro 
scope can be attached to the eye-piece of a telescope, so as to bring 
half the breadth of the solar spectrum side by side with half the 
breadth of the spectrum of the flame of some substance ; and their 
lines can thus be compared with each other on the divisions of the 
same scale. When this is done, there is found, with regard to 


several substances, an identity of position and relative breadth and 
intensity so exact that it is impossible to regard the agreement as 
accidental. The double line D of the sunbeam, is the prominent 
line of sodium. So all the numerous lines of potassium, iron, and 
several other simple substances, exactly coincide with the dark 
lines of the spectrum of sunlight. 

The foregoing facts seem to indicate that the photosphere of 
the sun consists of the flame of many substances, among which 
are some such as belong to the earth, namely, sodium, potassium, 
iron, &c. ; and that the luminous liquid matter beneath the pho 
tosphere shines through it, and changes all the bright lines to 
dark ones. 

633. Dispersion of Light. Decomposition of light refers to 
the fact of a separation of colors ; dispersion, rather to the meas 
ure or degree of that separation. The dispersive poiver of a me 
dium indicates the amount of separation which it produces, com 
pared with the amount of refraction. For example, if a substance, 
in refracting a beam of light 1 51 from its course, separates the 
violet from the red by 4 , then its dispersive power is T f T = .036. 
The following table gives the dispersive power of a few substances 
much used in optics : 

.Dispersive power. 

Oil of Cassia, . . 0.139 
Sulphuret of Carbon, 0.130 
Oil of Bitter Almonds, 0.079 
Flint-Glass, . . . 0.052 
Muriatic Acid, . . 0.043 
Diamond, .... 0.038 

Dispersive power. 

Plate-Glass, . . . 0.032 

Sulphuric Acid, . . 0.031 

Alcohol, .... 0.029 

Rock-Crystal, . . 0.026 

Blue Sapphire, . . 0.026 

Fluor-spar, . . . 0.022 

Crown-Glass, . . . 0.036 

The discovery that different substances produce different de 
grees of dispersion, is due to Dollond, who soon applied it to the 
removal of a serious difficulty in the construction of optical in 

634. Chromatic Aberration of Lenses. This is a devia 
tion of light from a focal point, occasioned by the different re- 
frangibility of the colors. If the surface of a lens be covered, 
except a narrow ring near the edge, and a sunbeam be transmitted 
through the ring, the chromatic aberration becomes very apparent ; 
for the most refrangible color, violet, comes to its focus nearest, 
and then the other colors in order, the focus of red being most 
remote. Since the distinctness of an image depends on the ac 
curate meeting of rays of the same pencil in one point, it is clear 
that discoloration and indistinctness are caused by the separation 
of colors. 



635. Achromatism. In order to refract light, and still keep 
the colors united, it is necessary that, after the beam has been 
refracted, and thus separated, a substance of greater dispersive 
power should be used, which may bring the colors together again, 
by refracting the beam only a part of the distance back to its 
original direction. For instance, suppose two prisms, one of 
crown-glass and one of flint-glass, each ground to such a refract 
ing angle as to separate the violet from the red ray by 4 . In 
order for this, the crown-glass, whose dispersive power is .036, 


must refract the beam 1 51 j for , = .036 ; and the flint- 

glass, whose dispersive power is .052, must refract only 1 17 ; 

f r -.o -, y/ = -052. Place these two prisms together, base to edge, 

as in Fig. 338, C being the crown-glass and F the flint-glass. Then 

FIG. 338. 

7 will refract the beam, II, downward 1 51 , and the violet, v, 4 
more than the red, r ; F will refract this decomposed beam up 
ward 1 17 , and the violet 4 more than the red, which will just 
bring them together at vr. Thus the colors are united again, and 
yet the beam is refracted downward 1 51 1 17 34 , from its 
original direction. 

636. Achromatic Lens. If two prisms can thus produce 
achromatism, the same may be effected by lenses ; for a convex 
lens of crown-glass may converge the rays of a pencil, and then a 


concave lens of flint-glass may diminish that convergency suffi 
ciently to unite the colors. A lens thus constructed of two lenses 


of different materials and opposite curvatures, so adapted as to 
produce an image free from chromatic aberration, is called an 
achromatic lens. Fig. 339 shows such a combination. The convex 
lens of crown-glass alone would gather the rays into a series of 
colored foci from v to r ; the concave flint-glass lens refracts them 
partly back again, and collects all the colors at one point, F. 

637. Colors not Dispersed Proportionally. It is assumed 
in the foregoing discussion, that when the red and violet are 
united, all the intermediate colors will be united also. It is found 
that this is not strictly true, but that different substances separate 
two given colors of the spectrum by intervals which have different 
ratios to the whole length of the spectrum. This departure from 
a constant ratio in the distances of the several colors, as dispersed 
by different media, is called the irrationality of dispersion. In 
consequence of it there will exist some slight discoloration in the 
image, after uniting the extreme colors. It is found better in 
practice to fit the curvatures of the lenses, for uniting those rays 
which most powerfully affect the eye. 



638. The Rainbow. This phenomenon, when exhibited 
most perfectly, consists of two colored circular arches, projected on 
falling rain, on which the sun is shining from the opposite part 
of the heavens. They are called the inner or primary bow, and 
the outer or secondary bow. Each contains all the colors of the 
spectrum, arranged in contrary order ; in the primary, red is out 
ermost ; in the secondary, violet is outermost. The primary bow 
is narrower and brighter than the secondary, and, when of unusual 
brightness, is accompanied by supernumerary bows, as they are 
called ; that is, narrow red arches just within it, or overlapping 
the violet ; sometimes three or four supernumeraries can be traced 
for a short distance. The common centre of the bows is in a line 
drawn from the sun through the eye of the spectator. 

639. Action of a Transparent Sphere on Light. It will 
aid in understanding the manner in which the bow is formed, to 
notice the experiments which first led to a correct theory respect 
ing it. Let a hollow sphere of glass be filled with water and placed 



FIG. 340. 

in the sunlight, and then let the directions and conditions of the 
most luminous pencils which emerge from it be observed. Fig. 
340 exhibits the general result. The entire hemisphere exposed 
to the rays is of course penetrated by them ; but a narrow pencil, 
S A, about 60 distant from S C, the axis 
of the drop, that is, the ray which passes 
through its centre, is converged to B, 
where some light escapes, but a large part 
is reflected to D ; at that point a division 
occurs again, and the emerging pencil, 
D E, consists of decomposed light, each 
color of which can be seen at a great dis 
tance. The part reflected at D divides 
again near F, but the emergent portion diverges, and has not the 
intensity of D E. But if a pencil, which strikes the drop at a 
about 10 outside of $ A, be traced, we find that a part is reflected 
near B, a part of that again reflected at d, and then, on reaching 
the point F, the emerging portion is not only decomposed, but re 
tains its intensity to a great distance. The pencil D E, which has 
been once reflected at B, is the one concerned in the production 
of the primary bow. The pencil F G, which leaves the drop at 
F, after it has been twice reflected, is one of those which consti 
tute the secondary bow. 

Most of the light which enters a drop of rain, and leaves it 
again, either not reflected at all, or reflected one or more times, is 
scattered in various directions, and brings to the eye of an ob 
server no impression of intense light. It is only such rays as are 
reflected and transmitted in circumstances to be contiguous to 
each other, and to continue parallel after leaving the drop, which 
can produce the bright colors of the rainbow. 

FIG. 341. 

640. Course of Rays in the Primary Bow. Let/ 2^2 
(Fig. 341) be the section of a drop of 
rain, fp a diameter, a ~b, c d, &c., par 
allel rays of the sun s light, falling 
upon the drop. Now yf, a ray coin 
ciding with the diameter, suffers no 
refraction ; and a b, a ray near to y f 9 
is refracted very little toward the ra 
dius, so as to meet the remoter surface 
of the drop about half as far from the 
axis as when it entered ; but the rays 
which lie further from yf 9 making greater angles with the radius, 
are more and more refracted as they are further removed from the 


And it is found, by a simple calculation on the course of the 
rays, that those which enter beyond the limit of about 60 J , cn> a 
more or less of those entering nearer the axis, the furthest one vT 
all at 90 being refracted almost to p. Hence all the rays fall in;; 
on the quadrant/ z, meet the circumference within the arc kp. 
But when a varying quantity is approaching its limit, or is begin 
ning to depart from it, its changes are nearly insensible. Thus, a 
large number of rays near c d y on both sides of it, meet very near 
k, the limit of the arc p k. Consequently, many more rays are re 
flected from that point than from any other in the arc. Now 
were these rays to return in the same lines, they would emerge 
parallel in the lines near c d ; but if, instead of returning back in 
the quadrant/ z, they are reflected on the other side of the radius, 
they make the same angles with the radius, and therefore with 
each other, as the incident rays do, and consequently meet the 
curve at the same inclination on the other side of the axis, and 
emerge parallel. Hence it appears that there is a particular point 
in the section of the drop on the back side, where the rays of the 
sun s light accumulate, and then diverge, so that, on emerging, 
those of a given color form a compact pencil of parallel rays. It 
is found by calculation that the angle which the incident and 
emergent rays make with each other that is, the angle included 
by cd and eq produced is, for the red rays, 42 2 , and for the 
violet rays, 40 17 , and for other colors, between these limits. Cal 
culation shows, also, that these are the greatest deviations possible 
for rays once reflected; since all rays on the quadrant /z, whether 
nearer or further than the pencil c d } at 60, emerge with smaller 

641. Course of Rays in the Secondary Bow. There is 
also an accumulation at a certain limit, for the light which 
emerges after suffering two reflections. If, as before, we calculate 
the course of the rays which fall on the quadrant /z, we shall find 
that those, d e (Fig. 342), which enter at about 71 or 72, from 
the axis fp, after crossing each other 
in the drop at , are reflected at h into 
parallel lines, and, consequently, after 
a second reflection at m, have their re 
lations to each other and the radii 
exactly reversed. Hence, they cross a 
second time at b, and emerge parallel 
at q. Such a pencil, entering above the 
axis, will, on emerging, ascend and 
cross its own path, outside of the drop, the violet rays intersecting 
d e at an angle of 54 9 , the red at 50 59 , and the other colors 

384 LIGHT. 

in order between. That the emergent pencil may descend to the 
observer, the incident pencil must enter Mow the axis, and come 
out above it. These rays, entering at the distance of 71 and 72 
from the axis, are the only ones which, after two reflections, emerge 
compact and parallel, and give a bright color at a great distance. 
All rays which enter nearer the axis, and also those which enter 
more remote, make, after two reflections, larger angles of devia 
tion, and also diverge from each other. 

642. Axis of the Bows. Let A B D G I (Fig. 343) repre 
sent the path of the pencil of red light in the primary bow. If 
A B and / G are produced to meet in K, the angle K is the devi 
ation, 42 2 , of the incident and emergent red rays. Suppose the 
spectator at /, and let a line from the sun be drawn through his 
position to T\ it is sensibly parallel to A B, and therefore the 
angles / and K are equal. As T is opposite to the sun, the red 

FIG. 344. 

color is seen at the distance of 42 2 , on the sky, from the point T\ 
and so the angular distance of each color from T equals the angle 
which the ray of that color makes with the incident ray. In like 
manner, in the secondary bow, if / T (Fig. 344) be drawn through 
the sun and the eye of the observer, it is parallel to A B, and the 
angular distance of the colored ray from T is equal to K, the devi 
ation of the incident and emergent rays. / T is called the axis of 
the bows, for a reason which is explained in the next article. 

643. Circular Form of the Bows. Let 8 O (Fig. 345) 
be a straight line passing from the sun, through the observer s 
place at 0, to the opposite point of the sky ; and let V 0, R be 
the extreme rays, which after one reflection bring colors to the eye 
at 0, and R 0, V O, those which exhibit colors after two reflec 
tions; then (according to Arts. 640, 641), VO C = 40 17 , ROC 
= 42 2 , R C= 50 59 , V O C = 54 9 . Now, if we sup 
pose the whole system of lines, S V O, S V 0, to revolve about 
S C, as an axis, the relations of the rays to the drops, and to 
each other will not be at all changed ; and the same colors will 
describe the same lines, whatever positions those lines may occupy 


ill the revolution. The emergent rays, therefore, all describe the 
surfaces of cones, whose common vertex is in the eye at ; and 

FIG. 345. 

the colors, as seen on the cloud, are the circumferences of their 

In a given position of the observer, the extent of the arches 
depends on the elevation of the sun. When on the horizon, the 
bows are semicircles ; but less as the sun is higher, because their 
centre is depressed as much below the horizon as the sun is ele 
vated above it. If rain is near, however, the lower parts of the 
bows may sometimes be seen projected on the landscape as arcs of 
ellipses, parabolas, or hyperbolas; for the surface of the earth cuts 
the axis of the cones obliquely. From the top of a mountain, the 
bows have been seen as entire circles. 

644. Colors of the Two Bows in Reversed Order. 

The reason for the inversion of colors in the two bows may be 
seen in the fact that, in the primary bow, the rays which descend 
to the observer s eye, must emerge from the lower or inner quad 
rant of the drop, and be bent upward (outward} from the radius 
produced; while, in the secondary, they must emerge from the 
upper or outer quadrant, and be bent from the radius downward. 
The ray V (Fig. 345), of the primary, being supposed a violet 
ray, is the most refrangible, and therefore all other rays from that 
drop fall below it, and fail to reach the eye. To bring other colors 
to 0) drops must be selected higher up ; hence, violet is the color 
seen nearest the axis. In the secondary bow, if V is the violet 
ray, the other colors, being bent in a less degree from the radius 
of the drop, lie above V 0; ainl therefore, in order that other 
colors may reach 0, they must emerge from lower drops, i. e. 
drops nearer the axis. Hence, violet is the outer color of the sec 
ondary bow. 


386 LIGHT. 

645. Rainbows, the Colored Borders of Illuminated 
Segments of the Sky. The primary bow is to be regarded as 
the outer edge of that part of the sky from which rays can come to 
the eye after suffering but one reflection in drops of rain ; and the 
secondary bow is the inner edge of that part from which light, 
after being twice reflected, can reach the eye. 

It is found by calculation, that in case of one reflection, the 
incident and emergent rays can make no inclinations with each 
other greater than 42 2 for red light, and 40 17 for violet; but 
the inclinations may be less in any degree down to 0. There 
fore, all light, once reflected, comes to the eye from within the 
primary bow. 

But the angles, 50 59 and 54 9 , are, by calculation, the least 
deviations of red and violet light from the incident rays after two 
reflections. But the deviations may be greater than these limits 
up to 180. Therefore rays twice reflected can come to the eye 
from any part of the sky, except between the secondary bow and 
its centre. 

It appears, then, that from the zone lying between the two 
bows, no light, reflected by drops internally, either once or twice, 
can possibly reach the eye. Observation confirms these state 
ments; when the bows are bright, the rain within the primary is 
more luminous than elsewhere; and outside of the secondary bow, 
there is more illumination than between the two bows, where the 
cloud is perceptibly darkest. 

646. The Tertiary Bow. A tertiary bow, or a bow formed 
by light three times reflected in drops of rain, is on the same side 
of the sky with the sun, and distant about 40 40 from it. The 
incident rays, which form it, enter the drops about 77 from their 
axis, and emerge on the back side. But this order of bow is so 
very faint from repeated reflections, and so unfavorably situated, 
that it is very rarely seen. 

647. The Common Halo. This, as usually seen, is a white 
or colored circle of about 22 radius, formed around the sun or 
moon. It might, without impropriety, be termed the frost-bow, 
since it is known to be formed by light refracted by crystals of ice 
suspended in the air. It is formed when the sun or moon shines 
through an atmosphere somewhat hazy. About the sun it is a 
white ring, with its inner edge red, and somewhat sharply defined, 
while its outer edge is colorless, and gradually shades off into the 
light of the sky. Around the moon it differs only in showing lit 
tle or no color on the inner edge. 

648. How Caused. The phenomenon is produced by light 



passing through crystals of ice, having sides inclined to each other 
at an angle of 00. Let the eye be at E (Fig. 346), and the sun in 
the direction E S. Let S A, S B, &c., be rays striking upon such 
crystals as may happen to lie in a position . 
to retract the light toward S E as an axis. 
Each crystal turns the ray from the re 
fracting edge on entering ; and again, on 
leaving, it is bent still more, and the emer 
gent pencil is decomposed. The color, which 
comes from each one to the eye E, depends 
on its angular distance from E S, and 
the position of its refracting angle. The 
angle of deviation for A is E A D=SJEJ A ; 
for By it is S E B, and so on. It is found 
by calculation, that the least deviation for 
red light is 21 45 ; the least for orange 
must be a little greater, because it is a lit 
tle more refrangible, and so on for the 
colors in order. The greatest deviation 
for the rays generally is about 43 13 . All 
light, therefore, which can be transmitted 
by such crystals must come to the observer from points some 
where between these two limits, 21 45 and 43 13 from the sun. 
But by far the greater part of it, as ascertained by calculation, 
passes through near the least limit. 

649. Its Circular Form. What takes place on one side of 
E S may occur on every side ; or, in other words, we may suppose 
the figure revolved about E S as an axis, and then the transmitted 
light will appear in a ring about the sun 8. The inner edge of 
the ring is red, since that color deviates least; just outside of the 
red the orange mingles with it; beyond that are the red, orange, 
and yellow combined ; and so on, till, at the minimum angle for 
Tiolet, all the colors will exist (though not in equal proportions), 
and the violet will be scarcely distinguishable from white. Beyond 
this narrow colored band the halo is white, growing more and 
more faint, so that its outer limit is not discernible at all. 

659. The Halo, a Bright Border cf an Illuminated Zone. 
As in the rainbow, so in the halo, the visible band of colors is 
only the border of a large illuminated space on the sky. The 
ordinary halo, therefore, is the bright inner border of a zone, which 
is more than 20 wide. The whole zone, except the inner edge, is 
too faint to be generally noticed, though it is perceptibly more 
luminous than the space between the halo and the luminary. 

338 LIGHT. 

651. Frequency of the Halo. The halo is less brilliant 
and beautiful, but far more frequent, than the rainbow. Scarcely 
a week passes during the whole year in which the phenomenon 
does not occur. In summer the crystals are three or four miles 
high, above the limit of perpetual frost. As the rainbow is some 
times seen in dew-drops on the ground, so the frost-bow, just after 
sunrise, has been noticed in the crystals which fringe the grass. 

652. The Mock Sun. The mock sun, or sun-dog, is a 
short arc of the halo, occasionally seen at 22 distance, on the 
right and left of the sun, when near the horizon. The crystals, 
which are concerned in producing the mock sun, are supposed to 
have the form of spiculce, or six-sided needles, whose alternate sides 
are inclined to each other at an angle of 60 ; these, if suspended 
in the air in a vertical position, could refract the light only in 
directions nearly horizontal, and therefore present only the right 
and left sides of the halo. 

In high latitudes, other and complex forms of halo are fre 
quent, depending for their formation on the prevalence of crystals 
of other angles than 60. [See Appendix for calculations of the 
angular radius of rainbows and halo.] 



653. Natural Colors of Bodies. The colors which bodies 
exhibit, when seen in ordinary white light, are owing to the fact 
that they decompose light by absorbing or transmitting some 
colors and reflecting the others. We say that a body has a certain 
color, whereas it only reflects that color ; a flower is called red, 
because it reflects only or principally red light ; another yellow, 
because it reflects yellow light, &c. A white surface is one which 
reflects all colors in their due proportion ; and such a surface, 
placed in the spectrum, assumes each color perfectly, since it is 
capable of reflecting all. A substance which reflects no light, or 
but very little, is black. What peculiarity of constitution that is 
which causes a substance to reflect a certain color, and to absorb 
others, is unknown. 

Very few objects have a color which exactly corresponds to any 
color of the spectrum. This is found to result from the fact that 


most bodies, while they reflect some one color chiefly, reflect the 
others in some degree. A red flower reflects the red light abun 
dantly, and perhaps some rays of all the other colors with the ivtl. 
Hence there may be as many shades of red as there can be differ 
ent proportions of other colors intermingled with it. The same 
is true of each color of the spectrum. Thus there is an infinite 
variety of tints in natural objects. These facts are readily estab 
lished by using the prism to decompose the light which bodies 

654. Inflection, or Diffraction of Light. This phenome 
non consists of delicate colored fringes bordering the edges of shad 
ows, when the light comes from a luminous point or line. 

For the purpose of experiments on this subject, a beam of light 
is admitted into a dark room, through a very small aperture, as a 
pin-hole made in sheet-lead ; or, what is better, a convex lens is 
placed in the window-shutter, which brings the rays to a focus, 
and affords a divergent pencil of light. If we introduce into this 
pencil any opaque body, as a knife-blade, for example, and observe 
the shadow which it casts on a white screen, we shall observe on 
both sides of the shadow fringes of colored light, the different col 
ors succeeding each other in the order of the spectrum, from violet 
to red. Three or four series can usually be discerned, the one 
nearest to the shadow being the most complete and distinct, and 
the remoter ones having fewer and fainter colors. The phenome 
non is independent of the density or thickness of the body which 
casts the shadow. The light, in passing by the edge or back of a 
knife, by a block of marble or a bubble of air in glass, is in each 
case affected in the same way. But if the body is very narrow, as, 
for example, a fine wire, a modification arises from the light which 
passes the opposite side ; for now fringes appear icitliin the shadow, 
and at a certain distance of the screen the central line of the 
shadow is the most luminous part of it. 

655. Breadth of Fringe varies with the Color. If, in 

the foregoing experiments, we use light of one color alone instead 
of white light, then the fringes are only of that color, separated 
from each other by lines which are comparatively dark ; and, on 
measuring the breadths and distances of fringes of different colors, 
those of red light are found to be widest, those of violet narrowest, 
and the other colors have breadths according to their order. This 
explains why, in the case of white light, the several colors appear 
in a series, with the red outermost ; for each element of the white 
light forms its own system of fringes, but the systems do not coin 
cide the wider ones project beyond the narrower, and thus be 
come separately visible. 

390 LIGHT. 

If the screen is moved further from the body, the distance of a 
given color from the edge of the shadow becomes greater, but not 
in proportion to the distance of the screen from the body; which 
proves that the color is not propagated in a straight line, but in a 
curve. These curves are found to be hyperbolas, having their con 
cavity on the side next the shadow, and are in fact a species of 

656. Light through Small Apertures. The phenomena of 
inflection are exhibited in a more interesting manner when we 
view with a magnifying glass a pencil of light after it has passed 
through a small aperture. For instance, in the cone already de 
scribed as radiating from the focus of a lens in a dark room, let a 
plate of lead be interposed, having a pin-hole pierced through it, 
and let the slender pencil of light which passes through the pin- 
hole fall on the magnifier. The aperture will be seen as a lumi 
nous circle surrounded by several rings, each consisting of a pris 
matic series. These are, in truth, the fringes formed by the edge 
of the circular puncture, but they are modified by the circum 
stance that the opposite edges are so near to each other. If, now, 
the plate be removed, and another interposed having two pin-holes, 
within one-eighth of an inch of each other, besides the colored 
rings round each, there is the additional phenomenon of long lines 
crossing the space between the apertures; the lines are nearly 
straight, and alternately luminous and dark, and varying in color, 
according to their distance from the central one. These lines are 
wholly due to the overlapping of two pencils of light, for on cov 
ering one of the apertures they entirely disappear. By combining 
circular apertures and narrow slits in various patterns in the screen 
of lead, very brilliant and beautiful effects are produced. 

657. "Why Inflection is not always noticed in looking 
by the Edges of Bodies. It must be understood that light is 
always inflected when it passes by the edges of bodies ; but that it 
is rarely observed, because, as light comes from various sources at 
once, the colors of each pencil are overlapped and reduced to 
whiteness by those of all the others. By using care to admit into 
the eye only isolated pencils of light, some cases of inflection may 
be observed which require no apparatus. If a person standing at 
some distance from a window holds close to his eye a book or 
other object having a straight edge, and passes it along so as to 
come into apparent coincidence with the sash -bars of the window, 
he will notice, when the edge of the book and the bar are very 
nearly in a range, that the latter is bordered with colors, the violefc 
extremity of the spectrum being on the side of the bar nearest to 


the book, and the red extremity on the other side. Again, the 
effect produced when light passes through a narrow aperture may 
be seen by looking at a distant lamp through the space between 
the bars of a pocket-rule, or between any two straight edges 
brought almost into contact. On each side of the lamp are seen 
several images of it, growing fainter with increased distance, and 
finely colored. An experiment still more interesting is to look at 
a distant lamp through the net-work of a bird s feather. There 
are several series of colored images, having a fixed arrangement in 
relation to the disposition of the minute apertures in the feather ; 
for the system of images revolves just as the feather itself is 

658. Striated Surfaces. If. the surface of any substance is 
ruled with line parallel grooves, 2000 or more to the inch, it will 
reflect bright colors when placed in the sunbeam. Mother-of- 
pearl and many kinds of sea-shell exhibit colors on account of 
delicate striae on their surface. These are the edges of thin lam 
ina? which compose the shell, and which crop out on the surface 
in fine and nearly parallel lines. It may be known that the color 
arises from such a cause, if, when the substance is impressed on 
fine cement, its colors are communicated to the cement. Indeed, 
it was in this way that Dr. Wollaston accidentally discovered the 
true cause of such colors. The changeable hues in the plumage 
of some birds, and. the wings of some insects, are owing to a 
striated structure of their surfaces. But the metals can be made 
to furnish the most brilliant spectra, by stamping them with steel 
dies, which have been first ruled by a diamond with lines from 
2000 to 10,000 per inch, and then hardened. Gilt buttons and 
other articles for dress are sometimes prepared in this manner, 
and are called iris ornaments. The color in a given case depends 
on the distance between the grooves, and the obliquity of the beam 
of light. Hence, the same surface, uniformly striated, may reflect 
all the colors, and every color many times, by a mere change in its 
inclination to the beam of light. 

659. Thin Laminae. Any transparent substance, when re 
duced in thickness to a few millionths of an inch, reflects brilliant 
colors, which vary with every change of thickness. Examples are 
seen in the thin laminae of air occupying cracks in glass and ice, 
and the interstices between plates of mica, also in thin films of oil 
on water, and alcohol on glass, but most remarkably in soapy 
water blown into very thin bubbles. 

If a lens of slight convexity is laid on a plane lens, and the 
two are pressed together by a screV, and viewed by reflected light, 

392 LIGHT. 

rings of color are seen arranged around the point of contact. The 
rings of least diameter are broadest and most brilliant, and each 
one contains the colors of the spectrum in their order, from violet 
on the inner edge to red on the outer. But the larger rings not 
only become narrower and paler, but contain fewer colors ; yet the 
succession is always in the same order as above. Increased pres 
sure causes the rings to dilate, while new ones start up at the 
centre, and enlarge also, until the centre becomes black, after 
which no new rings are formed. These are commonly called 
Newton s rings, because Sir Isaac Newton first investigated their 

660. Ratio of Thicknesses for Successive Rings. A 
given color appears in a circle around the point of contact, be 
cause equal thicknesses are thus arranged. If the diameters of 
the successive rings of any one color be carefully measured, their 
squares are found to be as the odd numbers, 1, 3, 5, 7 ; and hence 
the thicknesses of the laminae of air at the repetitions of the same 
color are as the same numbers. For, let Fig. 347 represent a sec 
tion of the spherical and plane 

surfaces in contact at a. Let FrG - 347 - 

a ^ ad, be the radii of two 

rings at their brightest points. 

Suppose a i, perpendicular to 

m n, to be produced till it 

meets the opposite point of 

the circle of which ag is an arc, and call that point/; then af is 

the diameter of the sphere of which the lens is a segment. Let 

b e, dg, be parallel to a i, and e li, g i, to m n, then we have 

(eh)" 2 : (giy : lah x lif:ai x if. 

But the distances between the two lenses being exceedingly 
small in comparison with the diameter of the sphere, lif and if 
may be taken as equal to /, whence, by substitution, 

(e h)* : (g iy . : a Ji x af : ai x af : : ah : ai * : b e : dg. 

Therefore the thicknesses of successive rings are as the odd 

661. Thickness of Laminae for Newton s Rings. The 

absolute thickness, be,dg, &c., can also be obtained, af being 
known, since 


for in so short arcs the chord may be considered equal to the sine, 
that is, the radius of the ring. When air is between the lenses, 
all the rings range between the thickness of half a millionth of an 


inch and 72 mUlionths ; if water is used, the limits are * of a mil 
lionth and 58 milliouths. Below the smaller limit the medium 
appears black, or no color is reflected ; above the highest limit 
the medium appears white, all colors being reflected together. 
AY hen water is substituted for air, all the rings contract in diam 
eter, indicating that a particular order of color requires less thick 
ness of water than of air ; the thicknesses for different media are 
found to be in the inverse ratio of the indices of refraction. 

662. Relation of Rings by Reflection and by Transmis 
sion. If the eye is placed beyond the lenses, the transmitted light 
also is seen to be arranged in very faint rings, the brightest por 
tions being at the same thicknesses as the darkest ones by reflec 
tion ; and these thicknesses are as the even numbers, 2, 4, G, &c. 
The centre, when black by reflection, is white by transmission, 
and where red appears on one side, blue is seen on the other; 
and, in like manner, each color by reflection answers to its comple 
mentary color by transmission. 

663. Newton s Rings by a Monochromatic Lamp. The 

number of reflected rings seen in common light is not usually 
greater than from five to ten. The number is thus small, because 
as the outer rings grow narrower by a more rapid separation of 
the surfaces, the different colors overlap each other, and produce 
whiteness. But if a light of only one color falls on the lenses, the 
number may be multiplied to several hundreds ; the rings are 
alternately of that color and black, growing more and more nar 
row at greater distances, till they can be traced only by a micro 
scope. A good light for such a purpose is the flame of an alcohol 
lamp, whose wick has been soaked in strong brine, and dried. 



664. Double Refraction. There are many transparent sub 
stances, particularly those of a crystalline structure, which, instead 
of refracting a beam of light in the ordinary mode, divide it into 
two learns. This effect is called double refraction, and substances 
which produce it are called doubly -refract ing substances. 

This phenomenon was first observed in a crystal of carbonate 
of lime, denominated Iceland spar. It is bounded by six rhom- 



FIG. 348. 

boidal faces, whose inclinations to each other are either 105 5 , or 
74 55 . There are two opposite solid angles, A and X (Fig. 348), 
each of which is formed by the meeting of three 
obtuse plane angles ; and when the edges of the 
crystal are equal, the diagonal A X is equally 
inclined to the edges which it meets, as A B, 
A C, and A D ; A X is called the axis of the 
crystal. But every other line in the crystal par 
allel to A X\& also an axis, because the crystal 
may be conceived to be divided into any number 
of similar crystals, each having its own axis ; the axis is therefore 
a direction rather than a line. If a thick crystal of spar be laid on 
a line of writing, it appears as two lines, one of which seems not 
only thrown aside from the other, but brought a little nearer to 
the eye. Therefore every ray of light, in passing through, is di 
vided into two rays, which come to the eye in different directions. 
The double refraction may also be seen by letting a very slender 
sunbeam, R r (Fig. 349), fall on the crystal ; as it enters it takes 
two directions, r 0, and r E, which on passing 
out describe the lines , E- E , parallel to 
the incident beam, R r. One of these rays, 
, is called the ordinary ray, because it is 
always refracted according to the ordinary law 
of refraction (Art. 606) ; that is, it remains in 
the plane of incidence, and the sines of inci 
dence and refraction have a constant ratio to 
each other at all inclinations. The other, E E 9 
is called the extraordinary ray, because in some 
positions it departs from this law of refraction 
in one or both particulars. 

The property of double refraction belongs to a large num 
ber of crystals, and also to some animal substances, as hair, 
quills, &c. ; and it may be produced artificially in glass by heat or 

665. Optical Relations of the Axis. The axis of Iceland 
spar has been defined with reference to form ; but it is also the 
axis with respect to its optical relations, for in the direction of 
that line a ray is never doubly refracted, while it is doubly re 
fracted in all other directions. 

Every plane which includes the axis of a crystal is called a 
principal section. In every principal section the extraordinary 
ray conforms to one part of the law of refraction, but not to the 
other ; it remains in the plane of incidence, but does not preserve 
a constant ratio of sines at different inclinations. 


In a plane at right angles to the axis, the extraordinary ray 
conforms to both parts of the law ; but in all planes besides this 
and the principal sections, it conforms to neither part. 

Crystals of a positive axis, are those in which the extraordinary 
ray lias a larger index of refraction than the ordinary ray ; crys 
tals of a negative axis are those in which the index of the extraor 
dinary ray is less than that of the ordinary ray. Iceland spar is a 
crystal of negative axis. 

Some crystals have two axes of double refraction ; that is, 
there are two directions in which light may be transmitted with 
out being doubly refracted. A few crystals have more than two 

666. Polarization of Light. This name is given to a 
change which may be produced in light, such that it has different 
properties on different sides. Common light, as, for instance, a di 
rect sunbeam, has the same relation to space on all sides. If it 
falls on a piece of glass at a given angle, it will suffer reflection 
equally well in every plane, as we turn the glass round, and so of 
refraction, or any change we may attempt. But if a beam were 
so changed in its character that it could be reflected upward, but 
could not be reflected to the right, it would be called, not common, 
but polarized light. 

667. Polarization by Reflection. Let two tubes, 
N P (Fig. 350), be fitted together in such a manner that one can 

FIG. 350. 

be revolved upon the other ; and to the end of each let there be 
attached a plate of dark-colored glass, A and (7, capable of reflect 
ing only from the first surface. These plates are hinged so as to 
be adjusted at any angle with the axis of the tube. Let the plane 
of each glass incline to the axis of the tube at an angle of 33, 
and let the beam R A make an incidence of 57, the complement 

396 LIGHT. 

of 33, on A ; then it will, after reflection, pass along the axis of 
the tube, and make the same angle of incidence on C. If now 
the tube NP be revolved, the second reflected ray will vary its 
intensity, according to the angle between the two planes of inci 
dence on A and C. The beam A C is polarized light ; the glass 
A, which has produced the polarization, is called the polarizing 
plate ; the glass (7, which shows, by the effects of its revolution, 
that A C is polarized, is the analyzing plate; and the whole in 
strument, constructed as here represented, or in any other manner 
for the same purpose, is called a polariscope. 

668. Changes of Intensity Described. The changes in 
the ray G E are as follows : When the tube N P is placed so that 
the plane of incidence on C is coincident with the former plane 
of incidence, R A C, whether GE is reflected forward or back 
ward in that plane, the intensity at E will be the same as if A C 
had been a beam of common light. If N P is revolved, E will 
begin to grow fainter, and reach its minimum of intensity when 
the planes RAG and A GE are at right angles, which is the 
position indicated in the figure. Continuing the revolution, we 
find the intensity increasing through the second quadrant of rev 
olution, and reaching its maximum, when the two planes of inci 
dence again coincide, 180 from the first position. The next half 
revolution repeats these changes in the same order. 

669. The Polarizing Angle. The angle of 57 is called the 
polarizing angle for glass, not because glass will not polarize at 
other angles of incidence, but because at all other angles it po 
larizes the light in a less degree ; and this is indicated by the fact 
that, in revolving the analyzing plate, there is less change of in- 
tsnsity, and the light at E does not become so faint. Different 
substances have different polarizing angles, and these are found to 
be so connected with the degree of refractive power, that by a 
knowledge of the index of refraction for any substance, its polar 
izing angle can be calculated, and vice versa. Hence the refractive 
power of opaque bodies may be determined. No substance en 
tirely polarizes the light incident upon it, even at the angle of 
polarization. Complete polarization of the ray A C would be 
indicated by the entire extinction of C E, at two opposite points 
of its revolution. On the other hand, every substance polarizes, 
in some degree, the light which it reflects. The polarization pro 
duced by reflection from the metals is very slight. 

670. Polarization by a Bundle of Plates. Light may 
also be polarized by transmission through a bundle of laminse of 


a transparent substance, at an angle of incidence equal to its po- 
larizing angle. Let a pile of twenty or thirty plates of transpa 
rent glass, no matter how thin, he placed in the same position as 
the reflector A, in Fig. 350, and a beam of light be transmitted 
through them in a direction toward ( . In entering and leaving 
the bundle A, situated as in the figure, the angles of incidence 
and refraction are in a horizontal plane. When O is revolved, the 
beam undergoes the same changes as before, with this difference, 
that the places of greatest and least intensity will be reversed. If 
the light is reflected from C in the same plane in which it was re 
fracted by .1, its intensity is least, and it is greatest when reflected 
in a plane at right angles to it, as at E in the iigure. 

671. Polarization by Crystals. The third and most per 
fect method of polarizing light, is by transmission through certain 
crystals. Some crystals polarize the transmitted light by absorp 
tion ; and every doubly-refracting crystal polarizes both the ordi 
nary and the extraordinary ray. If a thin plate be cut from a 
crystal of tourmaline, by planes parallel to its axis, the beam 
transmitted through it is polarized, and, when received on the 
analyzing plate, will alternately become bright and faint, as the 
tube of the analyzer is revolved. And if a beam is passed through 
a deubly-refracting crystal, and the two parts fall on the analyzing 
plate, they will come to their points of greatest and least bright 
ness at alternate quadrants ; indeed, when one ray is brightest, 
the other is entirely extinguished. Therefore the two rays which 
emerge from a doubly-refracting crystal are polarized completely, 
and in planes at right angles with each other. 

672. Every Polarizer an Analyzer. We have seen that 
light is polarized by reflection from glass at an incidence of 57, 
and analyzed by another plate at the same angle of incidence. 
This is but an instance of what is always true, that every method 
of polarizing light may be used to analyze, i. e., to test its polar 
ization. Hence, a bundle of thin plates of glass may take the 
place of the analyzer C\ as well as of the polarizer A. For, on 
turning it round, though the transmitted beam remains in the 
same place, yet it will, at the alternate quadrants, brighten to its 
maximum and fade to its minimum of intensity. 

So, again, if light has passed through a tourmaline, and is 
received on a second whose crystalline axis is parallel to that of 
the former, the ray will proceed through that also ; but if the 
second is turned in its own plane, the transmitted ray grows faint, 
and nearly disappears at the moment when the two axes are at 

398 LIGHT. 

90 of inclination, and this alternation continues at each 90 of 
the whole revolution. 

Finally, place a double-refractor at each end of the polari scope, 
and let a beam pass through them and fall on a screen. The first 
crystal will polarize each ray, and the second will doubly refract 
and also analyze each, exhibiting a very interesting series of 
changes. In general, four rays will emerge from the second crys 
tal, producing four luminous spots on the screen. But, on re 
volving the tube, not only do the rays commence a revolution 
round each other, but two of them increase in brightness, and the 
other two at the same time diminish as fast, till two alone are vis 
ible, at their greatest intensity. At the end of the second quad 
rant, the spots before invisible are at their maximum of bright 
ness, and the others are extinguished. This alternation continues 
as long as the crystal is revolved. In the middle of each quadrant 
the four are of equal brightness. 

673. Color by Polarized Light. The phenomena of color 
produced by polarized light are beautiful, and of great interest. 

Let a very thin plate of some doubly-refracting crystal be 
placed perpendicularly across the axis of the polariscope (Fig. 351), 
and let the analyzed ray, C E, fall on a screen. When the princi 
pal section of the crystal, D H, 
coincides with the first plane F IO- 351. 

of reflection, RAC,or is per 
pendicular to it, all the phe 
nomena are the same as if no 
crystal was interposed. But 
let the film be revolved in its 
own plane till D H makes 45 
with the plane EAC\ then, 

instead of the dark spot at E, a brilliant color appears. That 
color may be any tint of the spectrum, according to the thickness 
of the interposed film. If now the revolution of the crystal is 
continued, the color fades out at the end of the next 45, reap 
pears at 90, and so on. But if the crystal be so placed as to give 
color, and the analyzing plate be revolved, a different series pre 
sents itself. The color observed at E, during the first 45, grad 
ually fades, and during the next 45 its complement appears and 
brightens to its maximum. The original color is restored at 180, 
and the complementary color at 270. 

The most interesting form of this experiment is seen when the 
light is polarized and analyzed by means of double-refractors; 
since the polarization is more perfect, and the two pairs of oppo 
sitely polarized rays are on the screen at once. When two of the 


images are of a certain color, the other two have the complemen 
tary color. 

674. Systems of Colored Rings. Systems of iriscd Lands 
and rings may also be produced by the polariscope. Let a plaU 1 
be cut from a doubly-refracting crystal of one axis by planes per 
pendicular to that axis; and place it between the polarizer and 
analyzer. If now a pencil of sufficient divergency is transmitted, 
a system of colored circles will be formed, resembling Newton s 
rings between lenses. If a polariscope is formed of two tourma 
lines, and the crystal laid between them, and the whole combina 
tion, less than half an inch thick, is brought close to the eye, the 
pencil of light will consist of rays of various obliquity, and the 
rings may be seen beautifully projected on the sky. Or the ring 
systems may be projected on a screen by a polariscope furnished 
with concentrating lenses. Fig. 352 presents the system as seen 
through Iceland spar when the planes of reflection in the polari- 

FIQ. 352. FIG. 353. 

scope are at right angles. Two dark diameters cross the system 
and interrupt the rings. If the planes of reflection are coincident, 
the system is in every respect complementary to the other (Fig. 
353). The colors of the rings are all reversed, and the crossing 
bands are white. If double-refractors of two axes are used instead 
of the spar, compound systems are shown, of various forms and 
great beauty. 



675. The Wave Theory. Light has sometimes been re 
garded as consisting of material particles emanating from luminous 
bodies. But this, called the corpuscular or emission theory, has 
mostly yielded to the undulatory or wave theory, which supposes 

400 LIGHT. 

light to consist of vibrations in a medium. This medium, called 
the luminiferous ether, is imagined to exist throughout all space, 
and to be of such rarity as to pervade all other matter. It is sup 
posed also to be elastic in a very high degree, so that undulations 
excited in it are transmitted with great velocity. If radiant heat 
consists of undulations of the same ether, they perhaps differ from 
those of light only in being slower. For it is a familiar fact, that 
when the heat of a body is increased, a point is at length reached 
at which the body becomes luminous ; that is, the vibrations then 
affect the sense of sight as well as that of feeling. Moreover, the 
rays of heat are somewhat less refrangible than those of light (Art. 
635), from which it is inferred that its vibrations are slower. 

676. Postulates of the Wave Theory. 

1. The waves are propagated through the ether at the rate of 
192, 500 miles per second. 

As this is the known velocity of light, it must be the rate at 
which the waves are transmitted. 

2. The atoms of the ether vibrate at right angles to the line of 
the ray in all possible directions. 

It was at first assumed that the luminous vibrations, like the 
vibrations of sound, are longitudinal, that is, back and forth in the 
line of the ray ; but the discoveries in polarization require that the 
vibrations of light should be assumed to be transverse, that is, in 
a plane perpendicular to the line of the ray ; and, moreover, that 
in that plane the vibrations are in every possible direction within 
an inconceivably short space of time. Thus, if a person is looking 
at a star in the zenith, we must consider each atom of the ether 
between the star and his eye as vibrating across the vertical in all 
horizontal directions, north and south, east and west, and in innu 
merable lines between these. 

3. Different colors are caused by different rates of vibration. 
Red is caused by the slowest vibrations, and violet by the 

quickest, and other colors by intermediate rates. White light is to 
the eye what harmony is to the ear, the resultant effect of several 
rates of vibration combined. There are slower vibrations of the 
ether than those of red light, and quicker ones than those of violet 
light, but they are not adapted to affect the vision. The former 
affect the sense of feeling as heat, the latter produce chemical 
effects, and are called actinic rays. 

4. The ether within bodies is less elastic than in free space. 

This is inferred from the fact that light moves with less veloc 
ity in passing through bodies than in free space ; the greater the 
refractive power of a body, the slower does light move within it. 
And in some bodies of crystalline structure, it happens that the 



velocity is different in different directions, so that the elasticity of 
the ether within them must be regarded as varying with the 

677. Reflection and Refraction on the Wave Theory. 

When the waves of light reach the surface of a new medium, the 
ether within it being generally in a different state of elasticity, a 
system of waves will be propagated backward in the former me 
dium, and another onward in the new medium. The reflected 
system will make the same angle with the perpendicular as the 
incident system, analogous to the reflection of waves of water and 
of sound. But the system which enters the medium will change 
its direction according to its velocity in the medium ; and the 
velocity depends on the elasticity of the ether. Ri media of greater 
refractive power, the elasticity is considered to be less than in 
those of less refractive power ; and the waves are therefore propa 
gated more slowly in the former than in the latter. Let A B, d D 
(Fig. 354), represent the parallel waves of a beam falling on M N, 
the surface of a denser medium. The 
side of the wave which enters first at 
D, advances more slowly than the side 
still moving in the rarer medium. 
Suppose D to reach c, while d is going 
to 0\ then the wave, now wholly 
within the medium, lies in the posi 
tion C c, and advances in a line per 
pendicular to Cc, so long as it contin 
ues in the medium. Thus the light 
is refracted toward the perpendicular 
G H, in entering a denser medium. In a similar manner, it is 
shown that EFCc,iu entering a rarer medium, is refracted from 
the perpendicular, since the side C d emerges first and then gains 
velocity over the side c D. 

678. Refraction on the Emission Theory. It appears, 
therefore, that the wave theory requires us to suppose light to 
move more slowly in denser media. But, in the emission the 
ory, it is necessary to suppose it to move more swiftly. For the 
bending of the path of a particle of light toward the perpen 
dicular must be attributed to the attraction exerted by the 
medium on the particle. Suppose, then, that a particle of light 
moves along B A (Fig. 355), and enters a denser medium. Let 
the velocity, B A, be resolved into BE, E A\ the latter will be 
increased by the attraction of the medium ; the former will not be 
changed. Make A G = B E or D A, and A .F greater than A E; 
then A C represents the direction and velocity of the ray after 




FIG. 355. 

entering the medium. But as A F is greater than E A, while 
A G = D A, . . A C is greater than B A. On the other hand, if a 
ray, C A, is entering a rarer medium, the 
attraction of the denser draws it backward, 
and renders the component A E less than 
F A ; and hence the velocity A B, in the 
rarer medium, is less than C A, the velocity 
in the denser. The two theories are thus in 
conflict on the question whether light gains 
or loses velocity in entering a more refrac 
tive medium. Several direct tests have been 
applied in order to determine this ; and they 
all agree in proving that light moves more 
slowly in substances which have greater refractive power. 

679. Interference. Many interesting phenomena are ex 
plained on the principle of interference of waves. As two systems 
of water-waves may increase or diminish their height by being 
combined, and as sounds, when blended, may produce various 
results, and even destroy each other, so may two pencils of light 
either augment or diminish each other s brightness, and even pro 
duce darkness. 9 

Any one may try for himself the following experiment : Prick 
two very small holes, quite near each other, through paper, and 
holding the paper close to one eye, look through both holes at any 
small bright spot, such as occurs in a crack of glass when the sun 
shines upon it ; then will the bright spot be seen striped across 
with parallel black lines, which will be further apart as the holes 
are closer together. The two pencils of light, through the two 
apertures, overlap on the retina of the eye, and cause bright and 
dark lines by interference. Where like phases meet, the lines are 
bright ; where opposite phases meet, there is no light, and the lines 
are black. 

But there are other forms of experiment by which the exact 
length of wave for each color may be determined. 

680. Interference by Thin Plates. Let light of any one 
color, as yellow, fall on the lenses which exhibit Newton s rings. 
A system of waves is reflected from the first surface of the thin 
stratum of air which lies between the lenses, and another system 
from the second, and these two come to the eye together. Sup 
pose, at a given point, the thickness of air is such that the reflected 
waves of the second system meet those of the first, phase for phase, 
in exact concert ; at that point is seen a brighter yellow than if 
there was but one reflecting surface. But, at another point, the 


thickness of the air may be such that the two systems disagree by 
half a wave, bringing opposite phases together; in which case all 
motion is destroyed, and the point is black. The former is one 
point of a yellow circle, the latter of a black circle, each aror^ " 
the point of contact. It is obvious that at the smallest 1 
ring, the reflected waves from the second surface must be just one 
wave-length behind those from the first ; at the second ring, two 
waves behind, &c. ; and, in general, luminous circles appear where 
the two systems differ by an exact number of whole waves, and 
dark circles where they differ by half a wave, or any whole number 
and a half. The exact measurement of the thicknesses of air at 
any point (Art. 661), has led to the determination of the length of 
waves of each color. 

681. Change of Color alters the Size of the Rings. 

If orange or red light is used instead of yellow, the rings are a lit 
tle enlarged, being formed where the lamina of air is a little 
thicker, and therefore the waves for those colors are longer ; but 
if green, blue, indigo, and violet are each tried separately,- the 
rings grow smaller in each case ; and it is inferred that the lengths 
of waves are less in the same order, and in the ratio of the thick 

The reason becomes obvious why, in white light, the rings are 
few in number, and consist of a series of different colors, without 
any black circles between. As rings of different colors are of dif 
ferent sizes when separate, so when all colors are used together 
they will be arranged side by side, and some will be likely to fall 
where the black circles between others would occur. Again, as all 
the rings grow narrower at greater distances, because the thickness 
of the lamina increases faster, they crowd upon and overlap each 
other, and produce white light. Hence, a full prismatic series 
occurs only near the centre, and after five or ten repetitions, grow 
ing less and less perfect, white light covers the whole surface. 

682. Interference by Two Mirrors. If two plane reflec 
tors, inclined at a very obtuse angle, receive light from a minute 
radiant, and reflect it to one spot on a screen, the reflected pencils 
will interfere, and produce bright and dark lines. Suppose light 
of one color, as violet, flows from a radiant point A (Fig. 356) ; let 
mirrors B C and B D reflect it to the screen K L. F and E may 
be so selected that the ray A F + F G equals the ray A E + E G. 
Then G will be luminous, because the two paths being equal, the 
same phase of wave in each ray will occur at the point G. But if 
H be so situated that^l/+/// differs half a violet wave from 
A e + e H, then H will be a dark point, because opposite phases 



meet there. A similar point, /, will lie on the other side of G. 
Again, there are two points, K and L, one on each side of G, to 
each of which the whole path of light by one mirror will exceed 

FIG. 356. 

the whole by the other by just one violet wave ; those points are 
bright. Thus, there is a series of bright and dark points on the 
screen ; or rather a series of bright and dark hyperbolic lines, of 
which these points are sections. Other colors will give bands sep 
arated a little further, indicating longer waves. And white light, 
producing all these results at once, will give a repetition of the 
prismatic series. 

683. Interference by Inflection. One of the forms of 
inflection is explained as follows: Through an opaque screen, A B 
(Fig. 357), let there be a very narrow aperture, cd, by which is 
admitted the beam of light, efgli, of some 
one color, and emanating from a single point. 
That part of the aperture near d may be re 
garded as a luminous centre, from which em 
anate waves in all directions, and the same is 
true of the other part of the aperture near c. 
Let i be a point on one side of the beam, so 
situated that the distances d i and c i shall 
differ by half a wave of the color employed ; 
then, as opposite phases meet there, i will be 
a dark point. Let j be a point still further 
removed from the beam, where c j d j 
equals the length of a wave, then j will be 
luminous, since like phases meet in that 
point. This alternation will be repeated a few times till the lumi 
nous points become crowded and feeble. If the aperture is made 
narrower, the intervals Ji i, i /, &c., will increase, as they obviously 
must, in order to preserve c i d i equal to a half wave, and cj 
d j equal to a wave. Violet light produces the narrowest lines, 
red the widest, and white light the prismatic series, for the same 
reason as in Newton s rings. If A c, the left side of the screen, is 
entirely removed, so that light passes only one edge, d, the fringes 
will still exist, though somewhat modified. 



684. Length and Number of Luminous Waves. The 

other cases of inflection, and the phenomena of st-nation, as well 
as the supernumerary rainbows, are fully accounted for on the 
principle of interference. The careful measurements which have 
been made in nearly all these instances, have led, by so many in 
dependent methods, to the accurate determination of the length 
of a wave of each color. When the length of wave of any color 
is known, the number of vibrations per second is readily obtained 
by dividing the velocity of light by the length of the wave, ffhe 
remarkable results of these investigations are given in the follow 
ing table : 


Length of a wave 
in decimals of 
an inch. 

Number of vibrations per 

Extreme red, . . 



Orange, .... 
Yellow, .... 





577,000,000 ooo ooo 



62 2 ; OOO, OOO,OOO,OOO 

Indigo, .... 



600 ooo ooo ooo ooo 

Extreme violet, . . 



Mean, .... 



685. Change of Vibrations in Polarized Light It has 
been stated (Art. 676) that the vibrations of the ether, in the 
case of common light, must be supposed to be transverse in all 
directions. But, instead of this, we may conceive, what is me 
chanically equivalent to it, that the vibrations are made in two 
transverse directions at right angles to each other. Thus, in the 
descent of light from a star in the zenith, we may suppose each 
atom of the ether to vibrate in the two transverse lines, one north 
and south, and the other east and west ; because every motion 
oblique to these can be resolved into two components, one on 
each of these two. Or any other two lines, perpendicular to each 
other in the same plane, may be assumed as the directions of 

This being the nature of common light, it is easy to state 
what is meant by polarized light. It is that in which the vibra 
tions are performed in only one, of the transverse directions. For 
example, in the ray of star-light just supposed, if all the easterly 
and westerly vibrations, and all the easterly and westerly compo 
nents of the oblique vibrations, were destroyed, then no motions 

40$ LIGHT. 

would remain except in the north and south direction, and the 
light of that star would be polarized. It is, of course, immaterial 
what particular transverse motion is cut off, provided all the mo 
tion at right angles to it is retained. 

686. Polarizing and Analyzing by Reflection. When 
light is reflected, those vibrations of the ray which are in the 
plane of incidence are generally weakened in a greater or less de 
gree, while those which are perpendicular to the same plane are 
not affected. How much the vibrations are weakened depends on 
the elasticity of the ether withm the medium, and on the angle 
of incidence. But reflection of light rarely if ever takes place 
without diminishing the amplitude of those vibrations which are 
in the plane of incidence ; so that a reflected ray is always polar 
ized, at least, in a slight degree. 

It will now be readily understood how the analyzing plate 
(Fig. 350) proves the light to be polarized. Suppose the reflectors 
A and C are so perfect polarizers that vibration in the plane of 
incidence is entirely destroyed. Along R A the particles of ether 
vibrate across it both horizontally and vertically ; and as the plane 
of incidence R A C is horizontal, the atoms along A C will 
vibrate only vertically, because the horizontal vibrations, being 
in the plane of incidence, are destroyed. Now let C be placed so 
as to reflect horizontally ; the light will not be weakened by this 
reflection, because there are no horizontal vibrations to be de 
stroyed. But let C be turned so as to reflect vertically, for in 
stance, upward; now there can be no reflection, since all the 
vibrations left in A C are in the vertical plane, which is the plane 
of incidence ; and they are destroyed. For the same reason that 
reflection at A extinguished all horizontal motions in the atoms 
of ether, the reflection at C extinguishes all vertical motions; 
hence there is no motion beyond G. 

687. Polarizing by Transmission through a Bundle of 

Plates. At each of the surfaces some reflection occurs, so that 
all vibrations in the plane of incidence at length disappear from 
the reflected ray, even though the laminae are not perfect polar 
izers; while all vibrations perpendicular to this plane are pre 
served. Hence the reverse must be true of the transmitted ray ; 
it will retain the vibrations, so far as they coincide with the plane 
of incidence, and lose them, so far as they are perpendicular to it. 
Thus the two sets of rectangular vibrations are separated from 
each other ; one exists in the reflected ray, the other in the trans 
mitted ray. The two rays are therefore polarized in planes at 
right angles to each other. 


688. Polarizing by Absorption. A tourmaline absorbs, or 
in some way extinguishes the vibrations, so far as they are perpen 
dicular to its crystalline axis, but leaves all motion which is par 
allel to its axis unimpaired. It is at once apparent why a second 
tourmaline analyzes ; for if its axis is parallel to that of the first, 
the same vibrations which could pass the one, could pass the other 
also ; but if the two axes are at right angles, the same system of 
vibrations which could pass the first, because parallel to its axis, 
will be absorbed by the second, because perpendicular to its axis. 

689. Polarizing by Double Refraction. In doubly-re 
fracting crystals, the ether possesses different degrees of elasticity 
in different directions ; hence, so far as vibrations lie in one plane, 
they may be more retarded in their progress, and in a plane at 
right angles to that they may be less retarded, and the degree of 
refraction depends on the amount of retardation (Art. 677). 
Thus the two systems become separated, and emerge at different 
places. Each ray is of course polarized, having vibrations in only 
one direction ; and the two planes of polarization are at right 
angles to each other. 

690. Different Kinds of Polarization. Since the discovery 
was made that the etherial atoms may by certain methods be 
thrown into circular movements, and by others into vibrations in 
an ellipse with the axis in a fixed direction, the polarization 
already described has been called plane polarization, since the 
atoms vibrate in a plane. Circular polarization is that in which 
the atoms revolve in circles ; and elliptical polarization denotes a 
state of vibration in ellipses, whose major axes are confined to one 



691. Image by Light through an Aperture. If light from 
an external object pass through a small opening of any shape in 
the wall of a dark room, it will form an ill-defined inverted image 
on the opposite wall. Imagine a minute square orifice, through 
which the light enters and falls on a screen several feet distant. 
A pencil of light, in the shape of a* square pyramid, emanating from 
the highest point of the object, passes through the aperture, and 

408 LIGHT. 

forms a luminous square near the bottom of the screen. From an 
adjacent point another pencil, crossing the first at the aperture, 
forms another square, overlapping and nearly coinciding with the 
former. Thus every point of the object is represented by its 
square on the screen ; and as the pencils all cross at the aperture, 
the image formed is every way inverted. It is also indistinct, 
because the squares overlap, and the light of contiguous points is 
mingled together. If the orifice is smaller, the image is less lumi 
nous, but more distinct, because the pencils which form it overlap 
in a less degree. If the hole is circular, or triangular, or of irreg 
ular form, there is no change in the appearance of the image, 
which is now composed of small circles, or triangles, or irregular 
figures, whose shape is completely lost by overlapping. 

692. Effect of a Convex Lens at the Aperture. The 
image will become distinct, and more luminous also, if the aper 
ture be enlarged to a diameter of two or three inches, and then 
covered by a convex lens of the proper curvature. The image will 
be distinct, because the rays from each point of the object are con 
verged to a point again, and luminous, in proportion as the lens 
has a larger area than the aperture before employed. This is a 
real, and therefore an inverted image (Art. 618). A scioptic l)all 
is a sphere containing a lens, and so fitted in a socket that it can 
be turned in any direction, and thus bring into the room, the im 
ages of different parts of the landscape. The camera obscura is a 
darkened room furnished with a scioptic ball and adjustable screen 
for producing distinct pictures of external objects. 

Instead of connecting the lens with the wall of a room, it is 
frequently attached to a portable box or case, within which the 
image is formed. The Daguerreotype, or photograph, is the image 
produced by the convex lens, and rendered permanent by the 
chemical action of light on a surface properly prepared. The lens 
for photographic purposes needs to be achromatic, and corrected, 
also, as far as possible, for spherical aberration. 

693. The Eye. The eye is a camera obscura in miniature ; 
we find here the darkened room, the aperture, the convex lens, and 
the screen, with inverted images of external objects painted on it. 
A horizontal section of the eye is represented in Fig. 358. 

The optical apparatus of the eye, and the spherical case which 
incloses it, constitute what is called the eye-ball. The case itself, 
except about a sixth part of it in front, is a strong white substance, 
called, on account of its hardness, the sclerotic coat, 8, #(Fig. 358). 
In the front, this opaque coat changes to a perfectly transparent 
covering, called the cornea, C, C,. which is a little more convex than 
the sclerotic coat. The increased convexity of the cornea may be 



felt by laying the finger gently on the eye-lid when closed, and 
then rolling the eye 
one way and the other. 
The bony socket, 
which contains the 
eye, is of pyramidal 
form, its vertex being 
some distance behind 
the eye-ball; room is 
thus afforded for the 
mechanism which gives 
it motion. This cav 
ity, except the hemi 
sphere in front occu 
pied by the eye itself, 
is filled up with fatty 
matter and with the 
six muscles by which 
the eye-ball is revolved in all directions. 

694. The Interior of the Eye. Behind the cornea is a 
fluid, A, called the aqueous humor. In the back part of this fluid 
lies the iris, I, I, an opaque membrane, having in the centre of it 
a circular aperture, the pupil, through which the light enters. 
The iris is the colored part of the eye; the back side of it is black. 
Directly back of the aqueous humor and iris, is a flexible double 
convex lens, L, called the crystalline lens, or crystalline humor, 
having the greatest convexity on the back side. The large space 
back of the crystalline is occupied by the vitreous humor, V, a 
semi-liquid, of jelly-like consistency, Next to the vitreous humor 
succeed those inner coatings of the eye, which are most immedi 
ately concerned in vision. First in order is the retina. R, 7?, on 
which the light paints the inverted pictures of external objects. 
The fibres of the optic nerve, which enter the ball at N, are spread 
all over the retina, and convey the impressions produced there to 
the brain. Outside of the retina is the choroid coat, c h, c h, cov 
ered with a black pigment, which serves to absorb all the light so 
soon as it has passed through the retina and left its impressions. 
The choroid is inclosed by the sclerotic already described. The 
nerve-fibres, which are spread over the interior of the retina, are 
gathered into a compact bundle about one-tenth of an inch in 
diameter, which passes out through the three coatings at the back 
part of the ball, about fifteen degrees from the axis, X X, on the 
side toward the other eye. M, M represent two of the muscles, 
where they are attached to the eye-ball. 

410 LIGHT. 

695. Vision. The index of refraction for the cornea, and the 
aqueous and vitreous humors, is just about the same as that for 
water; for the crystalline lens, the index is a little greater. The 
light, therefore, which comes from without, is converged principally 
on entering the cornea, and this convergency is a little increased 
both on entering and leaving the crystalline. If the convergency 
is just sufficient to bring the rays of each pencil to a focus on the 
retina, then the images are perfectly formed, and there is distinct 
vision. To prevent the reflection of rays back and forth within 
the chamber of the eye, its walls are made perfectly black through 
out by a pigment which lines the choroid, the ciliary processes, 
and the back of the iris. Telescopes and other optical instru 
ments are painted black in the interior for a similar purpose. 

The cornea is prevented from producing spherical aberration 
by the form of a prolate spheroid which is given to its surface, and 
the crystalline, by a gradual increase of density ffom its edge to 
its centre. 

696. Adaptations. By the prominence of the cornea rays 
of considerable obliquity are converged into the pupil, so that the 
eye, without being turned, has a range of vision more or less per 
fect, through an angle of about 150. 

The quantity of light admitted into the eye is regulated by the 
size of the pupil. The iris, composed of a system of circular and 
radial muscles, expands or contracts the pupil according to the 
intensity of the light. These changes are involuntary ; a person 
may see them in his own eyes by shading them, and again letting 
a strong light fall upon them, while he is before a mirror. 

The pupils in the eyes of animals have different forms accord 
ing to their habits ; in the eyes of those which graze, the pupil is 
elongated horizontally, and in the eyes of beasts and birds of 
prey, it is elongated vertically. 

The eyes of animals are adapted, in respect to their refractive 
power, to the medium which surrounds them. Animals which 
inhabit the water have eyes which refract much more than those 
of land animals. The human eye being fitted for seeing in air, is 
unfit for distinct vision in water, since its refractive power is 
nearly the same as that of water, and therefore a pencil of parallel 
rays from water entering the eye would scarcely be converged at 
all. The effect is the same as if the cornea were deprived of all 
its convexity. 

697. Accommodation to Diminished Distance. It has 

been shown (Art. 618), that as an object approaches a lens, its 
image moves away, and the reverse. Therefore in the eye there 


must be some change in order to prevent this, and keep the image 
distinct 011 the retina while the object varies its distance. In a 
state of rest, the eye converges to the retina only the pencils of 
parallel rays, that is, those which come from objects at great dis 
tances. Rays from near objects diverge so much that, while the 
eye is at rest, it cannot sufficiently converge them so that they 
will meet on the retina ; but each conical pencil is cut off before 
reaching its focus, and all the points of the object are represented 
by overlapping circles, causing an indistinct image. The change 
in the eye, which fits it for seeing near objects distinctly, is called 
accommodation. This is effected by increasing the convexity of 
the crystalline lens, principally the front surface. The ciliary 
muscle, m, m, surrounds the crystalline, and is attached to the 
sclerotic coat just on the circle where it changes into the cornea. 
This muscle is connected with the edge of the crystalline by the 
circular ligament which surrounds the latter and holds it in place. 
When the muscle contracts, it relaxes the ligament, and the crys 
talline, by its own elastic force, begins to assume a more convex 
form, as represented by the dotted line. The eye is then accom 
modated for the vision of objects more or less near, 
according to the degree of change in the lens. On FlG - 859 - 
the other hand, when the ciliary muscle relaxes, the 
ligament again draws upon the lens to flatten it, and 
adapt it for the view of distant objects. In Fig. 359 
these two conditions of the crystalline are more dis 
tinctly shown. The dotted line exhibits the shape 
of the lens when accommodated for seeing near ob 
jects. Accompanying this action of the ciliary mus 
cle is that of the iris, which diminishes the pupil for 
near objects, so as to exclude the outer and more divergent rays. 
The dotted lines in front of the iris represent its situation when 
pushed forward by the crystalline accommodated for near objects. 

698. Long-Sightedness. As life advances, the crystalline 
becomes harder and less elastic. It therefore assumes a less con 
vex form when the ligament is relaxed, and cannot be accommo 
dated to so short distances as in earlier years ; and at length it 
remains so flattened in shape that only very distant objects can 
be seen distinctly. The eye is then said to be long-sighted, and 
requires a convex lens to be placed before it, to compensate for 
insufficient convexity in the crystalline. 

There are, however, cases of long-sightedness in early life. 
Such instances are found to be the result of an oblate form of the 
eye-ball, as shown in Fig. 3GO ; it is too short from front to back 
to furnish room for the convergency of the pencils, and they are 

412 LIGHT. 

cut off by the retina before reaching their focal points. In order 
to bring the distinct image forward upon the retina, convex 
glasses are needed in such 

cases, just as for the eyes FIG. 360. 

of most people when ad 
vanced in life. As the term 
long-sightedness is now ap 
plied to this abnormal con 
dition of the eye, the effect of 
age upon the sight is more 
properly called old-sightedness. 

699. Short - Sightedness. The eyes of the short-sighted 
have a form the reverse of that just described ; the eye-ball is 
elongated from cornea to retina (Fig. 361), resembling a prolate 
spheroid, so that rays parallel, or nearly so, are converged to a 
point before reaching the retina, and after crossing, fall on it in a 
circle ; and the image, made 

up of overlapping circles in- FlG - 

stead of points, is indistinct. 
If this elongation of the eye 
ball is extreme, an object 
must be brought very near, 
in order that its image may 
move back to the retina, and 
distinct vision be produced. 

This inconvenience is remedied by the use of concave lenses, 
which increase the divergency of the rays before they enter the 
eye, and thus throw their focal points further back. 

In the normal condition of the eyes in early life, the nearest 
limit of distinct vision is about five inches. This limit slowly 
increases with advance of life, but much more slowly in some 
cases than others, till it is at an indefinitely great distance. The 
near limit of distinct vision for the short-sighted varies from five 
down to two inches, according to the degree of elongation in the 

700. Why an Object is Seen Erect and Single. The 
image on the retina is inverted ; and that is the very reason why 
the object is seen erect ; the image is not the thing seen, but that 
~by means of which we see. The impression produced at any point 
on the retina is referred outward in a straight line through a point 
near the centre of the lens, to something external as its cause ; 
and therefore that is judged to be highest without us which makes 
its image lowest on the retina, and the reverse. 


An object appears as one, though we see it by means of two 
images ; but this is only one of many instances in which we have 
learned by experience to refer two or more sensations to one thing 
as the cause. Provided the images fall on parts of the retina, 
which in our ordinary vision correspond with each other, then by 
experience we refer both impressions to one object ; but if we 
press one eye aside, the image falls in a new place in relation to 
the other, and the object seems double. 

701. Indirect Vision. The Blind Point To obtain a 
clear and satisfactory view of an object, the axes of both eyes are 
turned directly upon it, in which case each image is at the centre 
of the retina. But when the light from an object is exceedingly 
faint, it is better seen by indirect vision, that is, by looking to a 
point a little on one side, and especially by changing the direction 
of the eyes from moment to moment, so that the image may fall 
in various places near the centre of the retina. Many heavenly 
bodies are plainly discerned by indirect vision, which are too faint 
to be seen by direct vision. 

In the description of the eye it was stated that the retina, as 
well as the choroid and the sclerotic, is perforated to allow the 
optic nerve to pass through. At that place there is no vision, and 
it is called the blind point. In each eye it is situated about 15 
from the centre of the retina toward the other eye. Let a person 
close his right eye, and with the left look at a small but conspicu 
ous object, and then slowly turn the eye away from it toward the 
right ; presently the object will entirely disappear, and as he looks 
still further to the right, it will after a moment reappear, and con 
tinue in sight till the axis of the eye is turned 70 or 80 from it. 
The same experiment may be tried with the right eye in the oppo 
site direction. The reason why people do not generally notice the 
fact till it is pointed out, is that an object cannot disappear to 
both eyes at once, nor to either eye alone, when directed to the 

702. Continuance of Impressions. The impression which 
a visible object makes upon the retina continues about one-eighth 
or one-ninth of a second ; so that if the object is removed for 
that length of time, and then occupies its place again, the vision 
is uninterrupted. A coal of fire whirled round a centre at the 
rate of eight or nine times per second, appears in all parts of the 
circumference at once. When riding in the cars, one sometimes 
gets a faint but apparently an uninterrupted view of the landscape 
beyond a board fence, by means of successive glimpses seen 
through the cracks between the upright boards. Two pictures, 

414 LIGHT. 

on opposite sides of a disk, are brought into view together, as 
parts of one and the same picture, by whirling the disk rapidly 
on one of its diameters. Such an instrument is called a tliauma- 
trope. The phantasmascope is constructed on the same principle. 
Several pictures are painted in the sectors of a circular disk, rep 
resenting the same object in a series of positions. These are 
viewed in a mirror through holes in the disk, as it revolves quickly 
in its own plane. Each glimpse which is caught whenever a hole 
conies before the eye, presents the object in a new attitude ; and 
all these views are in such rapid succession that they appear like 
one object going through the series of movements. 

703. Accidental Colors. There are impressions on the 
retina of another kind, which are produced by intense lights; they 
continue longer, and are in respect to color unlike the objects 
which cause them. They are commonly called accidental colors. 
If a particular part of the retina is for some time affected by the 
image of a bright colored object, and then the eyes are shut, or 
turned upon a white surface, the form appears to remain, but the 
color is complementary to that of the object ; and its continuance 
is for a few seconds or several minutes, according to the vividness 
of the impression. This is the cause of the green appearance of 
the sky between clouds of brilliant red in the morning or evening. 

704. Estimate of the Distance of Bodies. 

1. If objects are near, we judge of relative distance by the in 
clination of the optic axes to each other. The greater that incli 
nation is, or, which is the same thing, the greater the change of 
direction in an object, as it is viewed by one eye and then by the 
other, the nearer it is. If objects are ver y near, we can with one 
eye alone judge of their distance by the degree of effort required to 
accommodate the eye to that distance. 

2. If objects are known, we estimate their distance by the visual 
angle which they fill, having by experience learned to associate 
together their distance and their apparent, that is, their angular 

3. Our judgment of distant objects is influenced by their clear 
ness or obscurity. Mountains, and other features of a landscape, 
if seen for the first time when the air is remarkably pure, are esti 
mated by us nearer than they really are ; and the reverse, if the 
air is unusually hazy. 

4. Oar estimate of distance is more correct when many objects 
intervene. Hence it is that we are able to place that part of the 
sky which is near the horizon further from us than that which is 
over our heads. The apparent sky is not a hemisphere, but a flat 
tened semi-ellipsoid. 


705. Magnitude and Distance Associated. Our judg 
ments of distance and of magnitude are closely associated. If 
objects are known, we estimate their distance by their visual angle, 
as has been stated ; but if unknown, we must first acquire our 
notion of their distance by some other means, and then their visual 
angle gives us a definite impression as to their size. And if our 
judgment of distance is erroneous, a corresponding error attaches 
to our estimate of their magnitude. An insect crawling slowly 
on the window, if by mistake it is supposed to be some rods be 
yond the window, will appear like a bird flying in the air. The 
moon near the horizon seems larger than above us, because we are 
able to locate it at a greater distance. 

706. Binocular Vision. The Stereoscope. If objects are 
placed quite near us, we obtain simultaneously two views, which 
are essentially different from each other one with one eye, and 
one with the other. By the right eye more of the right side, and 
less of the left side, is seen, than by the left eye. Also, objects in 
the foreground fall further to the left compared witli distant ob 
jects, when seen with the right eye than when seen with the left. 
And we associate with th ese combined views the form and extent 
of a body, or group of bodies, particularly in respect to distance of 
parts from us. It is, then, by means of vision with two eyes, or 
binocular vision, that we are enabled to get accurate perceptions 
of prominence or depression of surface, reckoned in the visual 
direction. A picture offers no such advantage, since all its parts 
are on one surface, at a common distance from the eyes. But, if 
two perspective views of an object should be prepared, differing as 
those views do, which are seen by the two eyes, and if the right 
eye could then see only the right-hand view, and the left eye only 
the left-hand view, and if, furthermore, these two views could be 
made to appear on one and the same ground, the vision would 
then 1 be the same as is obtained of the real object by both eyes. 
This is effected by the stereoscope. Two photographic views are 
taken, in directions which make a small angle with each other, and 
these views are seen at once by the two eyes respectively, through 
a pair of half-lenses, placed with their thin edges toward each 
other, so as to turn the visual pencils away from each other, as 
though they emanated from one object. An appearance of relief 
and reality is thus given to superficial pictures, precisely like that 
obtained from viewing the objects themselves. 






707. The Camera Lucida. This is a four-sided prism, so 
contrived as to form an apparent image at a surface on which that 
image may be copied, the surface and image being both visible at 
the same time. It has the form represented by the section in Fig. 
362; A = 90, C= 135; B and D, 
of any convenient size, their sum of 
course = 135. A pencil of light 
from the object M, falling perpendic 
ularly on A D, proceeds on, and 
makes, with D C, an angle equal to 
the complement of D. After suffer 
ing total reflection at G, and again at 
H 9 its direction If E is perpendicular 
to MF. For, produce MF and EH, 
till they intersect in /; then, since 
(7= 135, CGH+CJIG = 5 ; but 
IGff=2CG, and IE G = 
2CHG-, . . IGH + IHG = $0 , 
/. / 90. Therefore HE emerges at right angles to A B, and is 
not refracted. Now, if the pupil of the eye be brought over the 
edge B, so that, while E H enters, there may also enter a pencil 
from the surface at M , then both the surface M and the object 
M will be seen coinciding with each other, and the hand may 
therefore sketch M on the surface at M r . The reason for two 
reflections of the light is, that the inversion produced by one 
reflection may be restored by the second. 

One of the most useful applications of the camera lucida is in 
connection with the compound microscope, where it is employed 
in copying with exactness the forms of natural objects, too small 
to be at all visible to the naked eye. 

708. The Microscope. This is an instrument for viewing 
minute objects. The nearer an object is brought to the eye, the 
larger is the angle which it fills, and therefore the more perfect is 
the view, provided the rays of each pencil are converged to a point 
on the retina. But if the object is nearer than the limit of dis 
tinct vision, the eye is unable to produce sufficient convergency. 
If the letters of a book are brought close to the eye, they become 
blurred and wholly illegible. But let a pin-hole be pricked through 


paper, and interposed between the eye and the letters, and, though 
lUint, they are distinct and much enlarged. The distinctness is 
owing to the fact that the outer rays, which are most divergent, 
are excluded, and the eye is able to converge the few central rays 
of each pencil to a focus. The letters appear magnified, because 
they are so near, and fill a large angle. The microscope utilizes 
these excluded rays, and renders the image not only large and dis 
tinct, but luminous. 

709. The Single Microscope. The single microscope is 
merely a convex lens. It aids the eye in converging the rays, 
which come from a very near object, so that a distinct and lumin 
ous image may be formed on the retina. The lens may be re 
garded as a part of the eye, and the diameter of an object is mag 
nified in the ratio of the limit of distinct vision to the focal 
distance of the lens. Taking five inches as the limit of distinct 
vision, if the principal focal distance is one-fourth of an inch, then 
we may consider the object twenty times nearer the eye than in 
viewing it without a lens, and therefore magnified twenty times in 
diameter, or 400 times in area. Now glass lenses are made whose 
focal length is not more than ^ inch, and whose magnifying 
power, therefore, is 5 : 5 $, = 250 in diameter, or 62,500 in area. 

Though the focal distance of a lens may be made as small as 
we please, yet a practical limit to the magnifying power is very 
soon reached. 

1. The field of view, that is, the extent of surface which can 
be seen at once, diminishes as the power is increased. 

2. Spherical aberration increases rapidly, because the outer 
rays are very divergent. Hence the necessity of diminishing the 
aperture of the lens, in order to exclude the most divergent rays. 

3. It is more difficult to illuminate the object as the focal 
length of the lens becomes less ; and this difficulty becomes a 
greater evil on account of the necessity of diminishing the aper 
ture in order to reduce the spherical aberration. 

Magnifying glasses are single microscopes of low power, such 
as are used by watchmakers. Lenses of still lower power and 
several inches in diameter are used for viewing pictures. 

710. The Compound Microscope. It is so called because 
it consists of two parts, an object-glass, by which a real and mag 
nified image is formed, and an eye-glass, by which that image is 
again magnified. Its general principle may be explained by Fig. 
363, in which a ~b is the small object, c d the object-glass, and ef 
the eye-glass. Let a b be a little beyond the principal focus of c d, 
and then the image g h will be real, on the opposite side of c d, 




and larger than a I. Now apply ef as a single microscope for 
viewing g h, as though it were an object of com 
paratively large size. Let g h be at the princi 
pal focus of ef, so that the rays of each pencil 
shall be parallel ; they will, therefore, come to 
the eye at &, from an apparent image on the 
same side as the real one, g h ; and the extreme 
pencils, eTc,flc 9 if produced backward, will in 
clude the image between them, e Icf being the 
angle which it fills. 

711. The Magnifying Power. The mag 
nifying power of the compound microscope is 

estimated by compounding two ratios ; first, the distance of the 
image from the object-glass, to the distance of the object from the 
same ; and secondly, the limit of distinct vision to the distance 
of the image from the eye-glass. For the image itself is enlarged 
in the first ratio (Art. 618) ; and the eye-glass enlarges that image 
in the second ratio (Art. 709). The advantage of this form over 
the single microscope is not so much that a great magnifying 
power is obtained, as that a given magnifying power is accom 
panied by a larger field of view. 

712. Modern Improvements. Great improvements have 
been made in the compound microscope, principally by combining 
lenses in such a manner as greatly to reduce the chromatic and 
spherical aberrations. The object-glass generally consists of one, 
two, or three achromatic pairs of lenses. The eye-piece usually 
contains two plano-convex lenses, a combination which is found 
to be the most favorable for diminishing the spherical aberration, 
and for enlarging the field of view. For convenience, the direc 
tion of the rays is, in many instruments, changed from a vertical 
to a horizontal direction, by total reflection in a right-angled 
prism. In Fig. 364, A is the object ; B, C, and C , achromatic 
piano - convex lenses, 

the piano - concave FlG - 

part being of flint- 
glass, the double-con 
vex part of crown- 
glass, and the two 
parts fitted and ce 
mented together; D 
the right-angled prism; E the field-glass, so called because it 
enlarges the field of view, by bending the outer pencils so that 
they come within the limit of the eye-glass G\ G the eye-glass, 



converging the pencils to the eye at //, while the rays of each 
pencil diverge a little, as from the magnified image back of 0. 
The image seen by the eye at H fills the angle I H L. 

713. Microscopes for Projecting Images. For the pur 
pose of forming magnified images on a screen, to be viewed by an 
audience, the microscope is modified in its arrangements. One 
form for projecting transparencies, whether paintings or photo 
graphs, is called the magic lantern. Another form, especially 
adapted for the exhibition of small objects in natural history, is 
the solar microscope. 

Such instruments are valuable as means of instruction and 
entertainment, but they are of no use for investigation and dis 

714. The Magic Lantern. It consists of a box, represented 
in Fig. 365, containing a lamp, and having openings so arranged 
as to permit the air to pass freely through it, without letting light 
escape. In front of the lamp is a tube containing a concentrating 
lens, C, the painting on glass, B 3 and the lens, A, for producing 

FIG. 365. 

the image ; back of the lamp may be a concave mirror for reflect 
ing additional light on the lens C. The transparency B is a 
painting on glass, and the strong light which falls on it proceeds 
through the lens A, as from an original object brilliantly colored. 
It is a little further from A than its principal focus, and therefore 
the rays from any point are converged to the conjugate focus in a 
real image, F, on a distant screen. This image is of course in 
verted relatively to the object, and therefore, if the picture B is 
inverted, F will be erect. The lens may be placed at various dis 
tances from B by the adjusting screw a, so as to give the greatest 
distinctness to the image at any given distance of the screen. Ac 
cording to Art. 618, the diam. of B : diam. of F : : A B : A F , and 
therefore, theoretically, the image may be as large as we please. 



But spherical aberration will increase rapidly as the image is en 
larged, and even if this evil could be remedied, the want of ligh t 
would render the image too faint to be well seen ; for the illumi 
nation is as much less than that of the painting as the area is 
greater. Two magic lanterns placed side by side, may throw dif 
ferent images on the same ground, so as to produce the effect 
called dissolving views. 

715. The Solar Microscope. This does not differ in prin 
ciple from the magic lantern. For illumination the solar or 
electric light is employed, and images are formed, not of artificial 
paintings, but of small natural objects. The lens A (Fig. 366), 

FIG. 366. 

which forms the image, is fixed in the end of a tube, A B, and at 
the other end is a mirror, M, which can be turned on a hinge to 
incline at any angle with the tube. This apparatus is attached to 
a window-shutter, the mirror on the outside, and the tube within. 
By adjusting screws the mirror is inclined so as to reflect the sun 
beam along the tube, where it is concentrated by lenses, L L, upon 
the object, 0. Just beyond the object is the lens A, of very small 
aperture, by which the image CD is formed. If the sunbeam is 
large, and the screen at a sufficient distance, the images of objects 
may be plainly seen when magnified millions of times in area. 
Spherical aberration, however, is considerable ; and this prevents 
the instrument from being of service for investigation. 

716. The Telescope. The telescope aids in mewing distant 
bodies. An image of the distant body is first formed in the prin 
cipal focus of a convex lens or a concave mirror; and then a 
microscope is employed to magnify that image as though it were 
a small body. The image is much more luminous than that 
formed in the eye, when looking at the heavenly body, because 
there is concentrated in the former the large beam of light which 



falls upon the lens or mirror, while the latter is formed by the 
slender pencil only which enters the pupil of the eye. If the 
image in a telescope is formed by a lens, the instrument is called 
a refracting telescope; but if by a mirror, a reflecting telescope. 

717. The Astronomical Telescope. This is the most 
simple of the refracting telescopes, consisting of a lens to form an 
image of the heavenly body, and a single microscope for magnify 
ing that image. The former is called the object-glass, the latter 
the eye-glass. The image is of course at the principal focus of 
the object-glass, and the eye-glass is placed at its own focal dis 
tance beyond the image, in order that the rays of each pencil may 
emerge parallel ; therefore the two lenses are separated from each 
other by the sum of their focal distances. The lines marked 
A, A , A" (Fig. 367), represent the cylinder of rays which flow 

FIG. 367. 

from the highest point of the object, and which cover the whole 
object-glass, M N. All these rays are collected at a, the lowest 
point of the image, the axis of the pencil, A a, being a straight 
line (Art. G15). After crossing at a, they are received on the 
lower edge of the eye-glass, P Q, by which they are made parallel, 
but the entire pencil is bent toward the axis of the lenses, and 
meets it at F. The beam, B, B , B", coming from the centre of 
the object, forms the centre, I, of the image ; and C, C , C", from 
the lowest point of the object, forms the top, c } of the image. In 
a similar manner each point of the image is formed by the con 
centrated rays which emanate from a corresponding point in the 
object. These innumerable pencils, after diverging from their 
focal points in the image, are turned toward the axis by passing 
through the eye-piece, while the rays of each become parallel. 
At F there is a diaphragm having an aperture, at which the eye 
is placed. 

718. The Powers of the Telescope. The magnifying 
power of the astronomical telescope is expressed by the ratio of the 
focal distance of the object-glass to that of the eye-glass. For (Fig. 

422 LIGHT. 

367) the object, as seen by the naked eye, fills the angle ADC, 
between the axes of its extreme pencils. But, since the axes cross 
each other in straight lines at the optic centre of the lens, ADO 
a D c. Therefore, to an eye placed at the object-glass, the 
image, a c, appears just as large as the object ; while at the eye 
glass it appears as much larger in diameter as the distance is less. 

The illuminating power is important for objects which shed a 
very feeble light on account of their immense distance. This 
power depends on the size of the beam, that is, on the aperture of 
the object-glass. 

The defining power is the power of giving a clear and sharply 
defined image, without which both the other pow r ers are useless. 
And it is the power of producing a well-defined image which lim 
its both of the other powers. For every attempt to increase the 
magnifying power by giving a large ratio to the focal lengths of 
the object-glass and the eye-glass, or to increase the illuminating 
power by enlarging the object-glass, increases the difficulties in 
the way of getting a perfect image. These difficulties are three 
the spherical aberration (Art. 621), the chromatic aberration (Art. 
634), and unequal densities in the glass. The third difficulty is a 
very serious one, especially in large lenses. Very few good object- 
glasses have been made so large as fifteen inches in diameter. 

719. Manner of Mounting. The equatorial mounting of 
large telescopes is quite essential for accuracy of observation or 
measurement. When the magnifying power is great, the diurnal 
motion is very perceptible, and the body quickly leaves the field 
of view. To prevent this, the telescope is so mounted as to re 
volve on an axis parallel to the earth s axis, and then by means of 
a clock it has a motion communicated to it, by which it exactly 
keeps up with the apparent motion of a heavenly body. Another 
axis, at right angles with the former, allows the telescope to be 
directed to a point at any distance north or south of the celestial 

Astronomical telescopes, when of portable size, are usually 
mounted upon a tripod stand, and admit of motion on a horizon 
tal and a vertical axis. 

720. The Terrestrial Telescope. In order to secure sim 
plicity, and thus the highest excellence, in the astronomical tele 
scope, the image is allowed to be inverted, which circumstance is 
of no importance in viewing heavenly bodies. But, for terrestrial 
objects, it would be a serious inconvenience ; and, therefore, a ter 
restrial telescope, or spy-glass, has additional lenses for the purpose 
of forming a second image, inverted, compared with the first, and, 
therefore, erect, compared with the object. In Fig. 368, m, m, m, 



represent a pencil of rays from the top of a distant object, and 
n, n, n, from the bottom ; A B, the object-glass ; nf n, the first 
image ; C D, the first eye-glass, which converges the pencils of 

FIG. 36& 

parallel rays to L. Instead of placing the eye at X, the pencils are 
allowed to cross and fall on the second eye-glass, E F, by which 
the rays of each pencil are converged to a point in the second 
image, m n , which is viewed by the third eye-glass, G H. The 
second and third lenses are commonly of equal focal length, and 
add nothing to the magnifying power. 

Such instruments are usually of a portable size, and hence the 
aberrations are corrected with comparative ease, by the methods 
already described. The spy-glass, for convenient transportation, 
is made of a series of tubes, which slide together in a very com 
pact form. 

721. Galileo s Telescope. This was the first form of tele 
scope, having been invented by Galileo, whose name it therefore 
bears. It differs from the common astronomical telescope in 
having for the eye-glass a concave instead of a convex lens, which 
receives the rays at such a distance from the focus to which they 
tend, as to render them parallel. Thus, the rays, M, M, M (Fig. 
369), from the top of the object, are converged by the object-glass, 

FIG. 369. 

A B, toward m, in the image; and the pencil, N, N, N 9 from the 
bottom of the object, is converged toward n ; but the concave lens 
CD is interposed at such a point as to render these converging 
rays parallel, and in this way they come to the eye situated behind 
the lens. But, though the rays converge before they reach the 
concave lens, the pencils diverge, having crossed at F\ therefore, 


in passing the concave lens, they are made to diverge more, and 
will enter the eye as if they had crossed at a much nearer point 
than F. The angle between these extreme pencils is the angle 
which the object appears to fill; and the magnifying power is in 
the ratio of this angle to the angle MFN=mFn- ) and that 
equals the ratio of the focal distance of A B to the focal distance 
of CD. The object appears erect in the Galilean telescope, since 
the pencil, which comes from the top of the object, appears to 
come from the top of the virtual image; thus, the parts of the 
object and image are similarly situated. It is obvious that, since 
the pencils diverge, only the central ones, within the size of the 
pupil, can enter the eye. This circumstance exceedingly limits 
the field of view, and unfits the instrument for telescopic use. It 
is employed for opera-glasses, having a power usually of only two 
or three in diameter. 

722. The Gregorian Telescope. This is the most frequent 
form of reflecting telescope, and receives its name from the inven 
tor, Dr. Gregory, of Scotland. The light from a heavenly body, 
entering the open tube (Fig. 370), is received on the large concave 

FIG. 370. 

speculum, E> which forms an inverted image, m, at the principal 
focus ; the rays of each pencil crossing there next meet the small 
concave mirror F, which forms an erect image, n, at the conjugate 
focus, beyond the speculum, the centre of the latter being perfo 
rated to let the light pass through. The eye-glass, G, magnifies 
this image. To avoid confusion, only two rays are drawn in the 
figure, and those belong to the central pencil. Eays from the top 
of the object would enter the tube inclining slightly downward, 
and be reflected to the bottom of m, and again to top of n. Eays 
from the bottom would ascend, and be reflected to the top of the 
first image, and to the bottom of the second. 

723. The Herschelian Telescope. Sir William Herschel 
modified the Gregorian by dispensing with the small reflector F 9 
and inclining the large speculum E, so as to form the image near 
the edge of the tube, where the eye-glass is attached. Thus, the 
observer is situated with his back to the object. The speculum of 
Herschel s telescope was about four feet in diameter, and weighed 
more than 2,000 pounds, and its focal length was forty feet. The 
Earl of Rosse has since constructed a Herschelian telescope having 
an aperture of six feet, and a focal length of fifty feet. 




1. Differential Equations for Force and Motion. These 
are three in number, as follows : 


L V = T( 

o r_^_^.f 

J ~ dt~ df 
3. fds = v dv. 

These equations are readily derived from the elementary prin- 


ciples of mechanics. In Art. 6 we have v = -. Eeducing the 


numerator and denominator to infinitesimals, v remains finite, and 

d s 
the equation becomes v = -3- ; which is Equation 1st. Therefore, 

if the space described by a body is regarded as a function of the 
time, the first differential coefficient expresses the velocity. 

Again (Art. 12), / = -, where / represents a constant force. 

Making velocity and time infinitely small, we get the intensity of 
the momentary force, /= ^-. But, by Equation 1st, v -^ ; 

d* s 
.*./ -j-j*9 which is Equation 2d. Hence we learn that the first 

differential coefficient of the velocity as a function of the time, or 
the second differential coefficient of the space as a function of the 
time, expresses the force. 

Equation 3d is obtained by multiplying the 1st and 2d cross 
wise, and removing the common denominator. 

We proceed to apply these equations to the preparation of for 
mula for falling bodies. 

2. Bodies falling through Small Distances near the 
Earth s Surface. In this case, let the accelerating force, which 



// W 

is considered constant, be called g. Then, by Eq. 2, g = ^-> .-. ^^ 
ydt. Integrating, we have v = gt +C. But, since v = 

when t = 0, .: v = gt, and = -, as in formulas 5, 6, Art. 28. 


Again, substituting g t for v in Eq. 1, dsgtdt\ and by 
integration, s = ^gt* + C\ but 6=0, for the same reason as be- 

fore ; . . s = ^gt\ and = y , as in formulas 1, 2, Art. 28. 


Once more, equating .the two foregoing values of t, we have 

_ , 

v = V2 g s, and s = ~-, as in formulas 3, 4, Art. 28. 

If, in the equation, s = ^ # f , v be substituted for # , we have 

s = %vt, or vt = 2s; that is, the acquired velocity multiplied by 
the time of fall gives a space twice as great as that fallen through 
(Art. 25). 

3. Bodies falling through Great Distances, 
so that G-ravity is Variable, according to the 
Law in Art. 16. 

Suppose a body to fall from A to B (Fig. 1), to 
ward the centre C. Let A C= a; B C = x; D C=r, 
the radius of the earth. 

The force/ at B, is found by the principle, Art. 16, 

FIG. 1. 

4. To find the Acquired Velocity. Substitute g r 9 x~* for 
/, and a x for s, in Equation 3d, and we have g r a ar 3 .d(a x) 
= vdv; /.by integration % v* = / g r 2 or 8 dx = g r a x~ l + C. 
But v = 0, when x = a ; /. C = g r*ar l ; and 


This is the general formula for the acquired velocity. If the 
body falls to the earth, x = r, and the formula becomes 

V = 



Again, if the body falls to the earth through so small a space 
that - may be regarded as a unit, the formula reduces to 

the same as obtained by other methods. 

If a body falls to the earth from an infinite distance, it does 
not acquire an infinite velocity. For then, as we may put a for 

(2 . 32| . 3956 . 5280)* feet = 6.95 miles. 

Therefore, the greatest possible velocity acquired in falling to 
the earth is less than seven miles ; and a body projected upward 
with that velocity would never return. 

5. To find the Time of Falling. From equation first we 
obtain d t ; in this, substitute d(a x) for d s, and j ^- 

for v, as found in the preceding article ; then 

i .1 i 

(d sc)* . d (a x) I ft- \* ^~ 


/. by integration t 

(~ J . / x* dx (a #) 

By the formula in the calculus for reducing the index of x we 

f- x^d x(a- x)~% = (ax- x^ - | vers~ l (^) + C. 

Now, when t = 0, x = a ; . . C = -^ ; 

hence , t m 

/ *\ 

(8 z _ a . )3 _ _ 

6. Bodies falling within the Earth (sup 
posed to be of uniform density), where 
Gravity Varies as the Distance from the 

Suppose a body to fall from A to B (Fig. 2) ; 
and let D C = r, A C = a, and B C = x. Then 

r:x::g:f=-x = force at B. 


To find the velocity acquired. By Eq. 3d, 

q 7 , x qxdx 
vdv fds\ . . vdv = -x.d (a x) = ^ ; 

/. i v*= -~ 1- C\ but v = when x = a\ 

q a? , t (a 3 # 2 ) $ <7 / * 2 \ 1 ~i 

.*. G = ^~, and ^ v = -^ 5 * v > ( ~" ^ ) f 

2 r > 2r (r v 

If the body falls from the surface to the centre, x = 0, and 

this formula becomes v = {fff$ = (32J x 3956 x 5280)^ = 25,904= 
feet per second. 

To find the time of falling. By Equation 1st, and substitu- 

ds d(ax) dx dx 

tions, we obtain at - 

V ^ ) (I / 9 *\ 

J */_ I /-/* _ // *\ 

\ r ( a x ) 


T = - cos- i -+a 

A w a 

When t = 0, a = , - = 1, and the arc, whose cosine is 1 = 0; 

"2 /> 

X COS" 1 -. 

/7*\"2" 77 

If the body falls to the centre, x 0, and t = (- j x ~ ; in 

^ c/ 

which a does not appear at all ; so that the time of falling to the 
centre from any "point within the surface is the same ; and equals 

73956 x ^ secons or 



7. Principle of Moments. In order to apply the processes 
of the calculus to the determination of the centre of gravity, the 
principle is used, which was proved (Art. 78), that if every par 
ticle of a body be multiplied by its distance from a plane, and 
the sum of the products be divided by the sum of the particles, 
the quotient is the distance of the common centre from the same 
plane. The product of any particle or body by its distance from 
the plane, is called its moment with respect to that plane. 

8. General Formulae. Let BAG (Fig. 3) be any symmetri 
cal curve, having A X for its axis of abscissas, and A Y, at right 



angles to it, for its axis of ordinates. It is obvious that the 
centre of gravity of the line B A C, of the area B A C, of 
the solid of revolution around the axis 
A X, and of the surface of the same FlG - 3 - 

solid, are all situated on A X, on ac 
count of the symmetry of the figure. 
It is proposed to find the formula for 
the distance of the centre from A Y, 
in each of these cases. Let G in every 
instance represent the* distance of the 
genera] centre of gravity from the axis 

A Y, or the plane A Y, at right angles to A X. The distance G 
would plainly be the same for the half figure B A D, as for the 
whole B A C; expressions may therefore be obtained for either, 
according to convenience. 

1. TJie line A B. Let x be the abscissa, and y the ordinate ; 

then (dx* + dy^ is the differential of the line A B. For brevity, 
let s = the line, and d s its differential. If we now multiply this 
differential by its distance from A Y, x d s is the moment of a 
minute portion of the line ; and the integral of it, / x d s, is the 
moment of the whole. Dividing this by the line itself, i. e. by s, 

we have 


for the distance G. 

2. Tlie area B A D. The differential of the area is y d x ; the 
differential of its moment is x y d x ; hence the moment itself is 

/?*?/ fJ *7* 

f x y d x ; and the distance G = - . 


3. TJte solid of revolution. The differential of the solid, gen 
erated by the revolution of A B on A X, is re y*d x ; the differen 
tial of its moment is rr x y*d x ; and the moment is / TT x y*d x ; 

hence the distance G = . 


4. The surface of revolution. The differential of the surface is 
2 rr y d s ; the differential of its moment is 2 re x y d s ; and there 
fore the moment isfZnxyds , and the distance G ^ir^ - 

J suriace 

9. Application of Formulae. We proceed to determine 
the centre of gravity in a few cases by the aid of these formula : 

1. A straight line. Imagine the line placed on A X, with one 
extremity at the origin A. The moment of a minute part of it is 
x d x, and that of the whole is f x d x, while the length of the 

whole is x ; .% G = 



= %x, as it evidently should 

430 LIGHT. 

be. In all the cases considered here, (7=0, because the function 
vanishes when x does. 

C x d s 

2. The arc of a circle. By formula 1st we have (7 = but 


ds = (dx > + difY ) by ^ ne equation of the circle, y* 2 a x x* ; 

( n /y\ a /7 /* // /y\ 2 /7 * 



fxds Px adx a T xdx a ( 

.-. = / - x = - / T =- \ vers l x 

s (2axx*}k s ^ (2ax x*)~ s ^ 

,~ , x -i ) a , ^ ail a c . ., ., 

(2 a x X*)? L = - (s y) = a a -, if the arc is dou- 

) 5 S t 

bled and called t, and c (chord) put for 2 y. Asa T is the dis- 


tance from the origin A, and a = radius of the arc ; .*. the distance 
from the centre of the circle to the centre of gravity of the arc, 

is ~, which is a fourth proportional to the arc, the chord, and the 


When the arc is a semi-circumference, c = 2 a, and t = Tia; 
/. the distance of the centre of gravity of a semi-circumference 

from the centre of the circle is . 


3. The area of a circular sector. Suppose the given sector to 
be divided into an infinite number of sectors ; then each may be 
considered a triangle, and its centre of gravity therefore distant 

O ff 

from the centre of the circle by the line -^-. Hence the centres of 


gravity of all the sectors lie in a circular arc, whose radius is ; 

so that the centre of gravity of the whole sector coincides 
the centre of gravity of that arc. The distance of the centre of 
gravity of the arc from the centre of the circle, by the preceding 

case, is~a x -c-4-- = - -, which is therefore the distance of 
o o o 6t 

the centre of gravity of the sector from the centre of the circle. 

When the sector is a semicircle the distance becomes - 

3 no, 


4. TJie area of a parabola. The equation of the curve is 

therefore the formula 2 for moment, 

13 15 

fx y d x = fp* x~ d x = \p~ a 
but the area of the half parabola = f p* x* 

o = x. 

To find the distance of the centre of gravity of the semi-parab 
ola from the axis A X, proceed as follows : The differential of the 
area, as before, equals y d x ; and the distance of its centre from 
A X is ^ y. Therefore its moment with respect to A X is A y * d x 
= p x dx\ and the moment of the whole is f^p x dx = \ p x 1 ; 
/. the distance of the centre from 

A X = \p x 9 -f- 3>M = f ^ x* = | y. 
5. TJie area of a circular segment. The equation of the circle 
is, y = (2 a x x*)^. Therefore (formula 2), 

fxydx =fx (2 ax -x^ dx. 

Add and subtract a (2 a x a;*) 2 d x, and it becomes 

fa (2 ax- x^dx -f(a - x) (2 a x x^dx = 
ax x*)* (a x) dx 

3 area A B& 

When x = a, G = a x ; and the distance of the centre of 

O 7T 

gravity of a semicircle from the centre of the circle = . When 

x = 2 a, G = a, as it plainly should be. 

6. A spherical segment. The equation of the circle is y* 
2 a x x 3 . Therefore (formula 3), 

fTTxy*dx=fn xdx(2ax-x*) =f2airx*dx 

_lanx* \rex* __%ax 

airx* IKX* ~~ 12a 4z* 

When x a, G f a ; that is, the centre of gravity of a hem 
isphere is g of radius from the surface, or | of radius from the 
centre of the sphere. If x = 2 a, G = a. 

7. A right cone. In this case A B (Fig. 3), is a straight line, 
and its equation is y = a x, where a is any constant 



= aV; . . fnxfdx = 

= V ;/.(? = = Jar. 

Hence the centre of gravity of a cone is three-fourths of the axis 
from the vertex. See Art. 75. 

8. The convex surface of a right cone. The equation is 

y=ax-, .-. dy* = tfdx*; and (dx* + dy*)% = (a? + l)?dx. 
Therefore (formula 4), 

= the moment of the surface. The surface itself, 

The centre of gravity of the convex surface of a right cone is on 
the axis, at a distance equal to two-thirds of its length from the 


9. To find the Moment of Inertia of a Body for any 

S (mr*) 
given Axis. To render the formula I = ~~Tfi7~ suitable to the 

application of the calculus, we have simply to substitute the sign 
of integration for S, and d M for m, and we have 



It is useful to know how to find the moment of inertia with respect 

to any axis by means of the FIG. 4. 

known moment with respect to 

another axis parallel to it and 

passing through the centre of 

gravity of the body. 

Let A Z (Fig. 4) be the axis 
passing through the centre of 
gravity of the body for which 
the moment of inertia is fr*dM, 
and let A Z be the axis paral 
lel to it, for which the moment 
of inertia, fr * d M of the same 
mass J/, is to be determined. 
For every particle m of the body 
the corresponding value of A m 
is r a = x* + y\ In like man- 


ner, if we denote the co-ordinates of A by a and j3, and the dis 
tance between the axes by a, we shall have a a = a a + /3 3 . Now the 
distance of the particle m from A Z is r 2 = (x a) 2 + (y /3) 8 

a 2 - Z ax 2 (3 y; .\ 

+ fr*dM,. ........ (2) 

since A Z passes through the centre of gravity of the body. Hence, 
the moment of inertia of a body with respect to any axis is equal to 
the moment of inertia with respect to a parallel axis through the 
centre of gravity, plus the mass of the body multiplied by the square 
of the distance between the two axes. 

Put C the moment of inertia with respect to an axis through 
the centre of gravity ; then the distance from the axis of suspen 
sion to the centre of oscillation, the axes being parallel, will be 

10. Examples. 

1. Find the centre of oscillation of a slender rod or straight 
line suspended at any point. 

Let a and b be the lengths on opposite sides of the axis of sus 
pension, then by (1) 

- fr dM__ fr dr 2 (a 3 .+ ft ) 2 (a* - a b + V] 

Mk ~ (a + b)%(a- b)~3(a* - b*)~ 3 (a - b) 

between the limits r = + a and r = b. 

If the rod is suspended at its extremity, b = 0, and I = f a. If 
it is suspended at its middle point, a = b and I = QC . 

2. Find the centre of oscillation of an isosceles triangle vibra 
ting about an axis in its own plane passing through its vertex. 

Put b and h for the base and altitude of the triangle; then by 



, b . 
r . r d r 

If the axis of suspension coincides with the base of the trian- 

r r \\ l(h-r)dr 
, ,, , i/o A v h 

g lc,thenl= , bh ^ h = g . 

3. Find the centre of oscillation of a circle vibrating about an 
axis in its own plane. 

C =fr* dM= 2fx* ydx = 2./V (1? - x*)? dx = 


Taking this integral between x = r and x = -f r, we have 

2*2 4 

Substituting this value of C in (3) we have 

^ + a 9 TT R* 


4. Find the centre of oscillation of a cir 
cle vibrating about an axis perpendicular 
to it. 

Let K L (Fig. 5) be an elementary ring 
whose radius is x and whose breadth is d x ; 

dM=27Txdx, and 0= I x* . 2nx d x 


FIG. 5. 

7T.fi 4 


7? 2 

As 4- TT- is greater than a + ^ , a cir- 
2 4 a 

cular pendulum will vibrate faster when the 
axis of suspension is in its plane, than when 
it is perpendicular to it. 


11. General Formula. Let the surface pressed upon be 
plane and vertical ; and let the water level be the plane of refer 
ence. Suppose the surface to have a 
symmetrical form with reference to a 
vertical axis, x, whose ordinate is y 
(Fig. 6). A horizontal element of the 
surface is 2 y d x, and (since the pres 
sure varies as the depth) the pressure 
on that element 2 x y d x. Hence the 
whole pressure to the depth x is 
f2xydx=2fzydx. The mo 
ment of the pressure on the element 

of surface is 2 x* y d x\ and the sum of all the moments to the 
same depth is / 2 x* y d x = 2 f x* y d x. Therefore, putting p 

C x* 11 d x 

for the depth of the centre of pressure,^ = -^- 7 -. 

f xy d x 


12. Examples. 

1. A rectangle. Let its height = //, and its base = b ; then 2 y 
everywhere equals #, and a horizontal element at the depth x is 
b d x, the pressure on it is b x d x, and the moment of that 
pressure is b x* d x ; .-. the depth of the centre of pressure p = 

f I x 9 d x J b x 3 + c 

r^ T = T-r-rT"1/ Since the pressure and area is each zero, 

J b x d x -j b x* -f c 

when x is zero, c and c both disappear, and p = f x, which for the 
whole surface becomes p h. That is, the centre of pressure on 
a vertical rectangular surface reaching to the water level, is two- 
thirds of the distance from the middle of the upper side to the 
middle of the lower. 

2. A triangle whose vertex is at the surface of the tvater, and its 
base horizontal. Let the triangle be isosceles, its height = h, and 

, J ^ 

its base = 1; then h : b : : x : 2 y = T x. Therefore^ = -7^ 



= y-l = j x ; and for the whole height, | h. 

3 X 

If the triangle is not isosceles, it may be easily shown that the 
centre of pressure is on the line joining the vertex and the middle 
of the base, at a distance from the vertex equal to three-fourths of 
the length of that line. 

3. A triangle whose base is at the ivater level. Then h : b 

: : h x : 2 y = b Y x Therefore the pressure is f b x d x 

f j x* d x, because d x is negative. The moment of the 
pressure is / b x* d x f -= x*d x. 

- x* +r 
Therefore p = 

x* 3 x* 

s - ; and, when x = h, this becomes .1 h. 

6 h x 4 x b h 4# 

In general, the centre of pressure is at the middle of the line join 
ing the vertex and the middle of the base. 

4. A parabola whose vertex is at the surface. As y = p 2 x*, 

fx*p?x l dx fx*dx %x* 5 . 5 . 
therefore p = - - = I - = ~z; or = h, for 

fx&&dx fx*dz fa* 

the whole area. 



5. A parabola whose base is at the surface. As li x is the 

/* / 7 \ 2 ^7 ^J 

depth of an element, d x is negative, p - = 

r Ch _ /y\ ^ fl n* 

f (h* x? d x - 2 h x% d x + a? dx) _ | / 

f(h-x) x 

- j h x* + f aff 

f(h x*dx-x*dz) " f 

- - ; and when x = 7^, the expression becomes 

iJlzJ * +** * &. 

i h in, 7 



13. The Primary Rainbow. Since the primary bow is 
formed by those rays which, on emerging after one reflection, 
make the largest angle with the incident rays, proceed to find 
what angle of incidence will cause the largest deviation of the 
emerging rays. 

In Fig. 7, let x = angle of inci 
dence ; y = angle of refraction ; z = _ 
angle of deviation ; n = index of re 
fraction. Then, in the quadrilateral 
B DG K, DBK=DGK=x-y, 
angle at D = 360 - 2 y ; .-. K z = 

FIG. 7. 

. dz 
. -^ 

A \j* 



sin x = n sin y\ 

.. , dy cos x 
/. cos x dx = n cos y dy, and 

(JL X 

By substitution. 

4 cos 

n cos y 

= 2. 

ncos y 

.: 2 cos x = n cos y ; and 4 cos 9 x n* cos 2 
But sin 2 x = M? sin 2 y ; 

.*. 3 cos 2 x + 1 = ri* ; since sin 3 + cos 2 = 1. 

.*. COS X = 

f - 1 

If 1.33 and 1.55, the values of n for extreme red and violet, be 
used in this formula, we obtain x, and therefore y and z, for the 
limiting angles of the primary bow. 



FIG. 8. 


14. The Secondary Bow. To find the angle of minimum 
deviation. Using the same notation as before, we have in tin- 
pentagon G EDBK (Fig. 8), G = II = 

18Q x + y; ED-^y\ .-. K=z 
180 + 2x 6y; 

d z G d 11 

" dx~ dx 

6 cos x 

.-. - = 2 ; and 3 cos x = n cos y ; 
ncos y 

.\ 9 cos 2 x = n* cos 2 y ; 
but sin 2 x = n* sin 2 y ; 

/. 8 cos 2 x + 1 = * ; 

/. cos a; = 

which, as before, will furnish z for each limiting color of the sec 
ondary bow. 

15. The Common Halo. Let D E (Fig. 9) be the ray from 
the sun, and F G the emergent ray. Let D Ep x\ K E F y ; 

FIG. 9. 

.-. z - x + y C f . 
sin x = n sin y, 
and sin y = n sin # ; 

/. x = sin" 1 (n sin y\ 
and y = sin" 1 (n sin # ) = 

sin" 1 \n sin (7 #}, 
By substitution, 

z sin" 1 (n sin ?/) + sin" 1 \n sin (G y)\ C. Therefore z is a 
function of # ; and, by differentiating, we have 
d z _ n cos y n cos (G y) 

dy~ Vl tflsijtf" 
n 3 cos 2 V 

x y + y % 

YL n snr (U y 

rc a cos 2 (C-y) 

1 ft 2 sin 2 ^ ~~ 1 ?i 2 sin 2 (6 r y) 

~ sn y _ 


* 1 - w a m 2 y ~~ 1 - n 1 sin 2 (C - y) 9 
.-. (rc 2 - 1) sin 2 y = (?i 2 - 1) sin 2 (C - y); 
.-. y = C - y, and y = J (7; 
and a/ = (7. 

Hence, the minimum deviation occurs when the ray within the 
crystal is equally inclined to the sides. Knowing n, the index of 
refraction for ice, x, and its ecp-i!. /, can be obtained, and then z, 
the deviation required. 

t *vt i,